Comparative Effectivenes of Newer Medications for Lower Urinary Tract Symptoms Attributed to Benign Prostatic Hyperplasia: A Systematic Review and Meta-analysis

2.1 Protocol

We developed an a priori written protocol, together with a technical report that incorporated input from key stakeholders, a multidisciplinary Technical Expert Panel, and public comment (available at the Agency for Healthcare Research and Quality [AHRQ] website http://effectivehealthcare.ahrq.gov/index.cfm/search-for-guides-reviews-and-reports/?productid=2067&pageaction=displayproduct ).

2.2 Eligibility criteria

Based on our Population, Interventions, Comparisons, Outcomes, Timing, and Setting criteria (Supplementary data) we included randomized controlled trials (RCTs) that tested comparative effectiveness of treatments involving newer drugs in men aged ≥45 yr with LUTS attributed to BPH. We defined these newer drugs as those that were Food and Drug Administration (FDA) approved for BPH since 2008 or which, though not FDA approved for BPH, have been studied for the treatment of BPH since 2008 and were selected through a formal process of stakeholder involvement (Supplementary data). Comparators included medications FDA approved for BPH before 2008. Included RCTs were at least 1 mo in duration with no minimum sample size. We additionally searched for large ( n ≥ 100 patients), longer-term (≥ 1 yr duration) observational studies to assess long-term or rare treatment associated harms only. We limited inclusion to English language articles.

The primary predefined outcomes of interest were changes reflecting clinically important differences (Supplementary data) in validated measures to assess LUTS (International Prostate Symptom Score [I-PSS]: score ranges 0–35 with higher scores indicating more severe symptoms; or American Urological Association Symptom Index scores), prostate-related bother or quality of life (QoL; I-PSS QoL question; BPH/LUTS impact scale), as well as rates of disease progression and/or treatment failure (prevention/delay of need for surgical intervention; acute urinary retention [AUR]). We also assessed common and serious medication adverse effects (AEs).

2.3 Information sources and literature search

We searched Ovid Medline, Ovid Embase, and the Cochrane Central Register of Controlled Trials with filters for study design (Supplementary data), to identify relevant RCTs published through June 20, 2016. We also searched for relevant systematic reviews and other key references. Lastly, we searched the Clinical Trials ( www.clinicaltrials.gov ) and the FDA ( www.fda.gov/Drugs ) websites to identify additional completed and ongoing studies.

2.4 Study selection process, data extraction, and risk of bias in studies

Two independent investigators screened titles and abstracts to identify studies meeting the eligibility criteria. Data were extracted by one investigator and reviewed and verified for accuracy by a second investigator. Risk of bias (RoB) of eligible studies was assessed using AHRQ guidance by one investigator and reviewed by a second [7] .

2.5 Synthesis of results

We assessed clinical and methodological heterogeneity and variation in effect size to determine the appropriateness of pooling data [8] . When three or more trials reported similar comparisons and outcomes, data were pooled using a Hartung, Knapp, Sidik, and Jonkman method [9] random effects model for proportion of I-PSS responders or mean changes in I-PSS scores in Stata (StataCorp., College Station, TX, USA). We pooled other outcomes in RevMan (RevMan, Spartanburg, SC, USA) [10] and converted DerSimonian-Laird random effects confidence intervals to Hartung, Knapp, Sidik, and Jonkman confidence intervals using an excel spreadsheet provided in Inthout et al [9] . We assessed between study variance with Tau ² and measured the magnitude of heterogeneity with the I ² statistic. If substantial heterogeneity was present (ie, I ² ≥ 70%), we stratified the results to assess treatment effects based on patient or study characteristics and/or explored sensitivity analyses [8 11] . We pooled across different ABs unless there were at least three trials for a given agent. We interpreted efficacy and comparative effectiveness using established thresholds indicating clinical significance (Supplementary data).

For the body of evidence from RCTs, we rated our confidence in the estimates of effect for the primary outcomes as overall strength of evidence (SOE) as high, moderate, low, or insufficient (Supplementary data) [12] . For observational studies, we did not formally assess SOE, but provided descriptive information in narrative form.

3.1 Search results

Our literature search identified 1139 references, of which 124 were selected for full-text review ( Fig. 1 ). This process mapped to 43 unique RCTs. In addition, we identified five longer duration (≥ 1 yr) observational studies reporting AEs.

 

Fig. 1
Figure 1: Literature flow diagram. The results are presented separately for each of four drug classes (new alpha-blockers, anticholinergics, beta-3 agonists, and phosphodiesterase type-5 inhibitor), and specific drugs are listed within each class. The outcomes addressed by the three key questions are discussed within each drug-specific section.

BPH = benign prostatic hyperplasia; RCT = randomized controlled trial.

3.2 Silodosin versus tamsulosin

Ten trials randomized men with LUTS attributed to BPH ( n = 1799) to silodosin 8 mg daily versus tamsulosin 0.2–0.4 mg daily 13 14 15 16 17 18 19 20 21 22 . Table 1 provides baseline characteristics. Overall, the RoB was low in two trials [13 18] , moderate in six trials 14 15 16 20 21 22 , and high in two trials [17 19] .

Three trials conducted responder analyses (defined as > 25% reduction in I-PSS score; Table 2 ) [16 18 20] . Response to treatment was not significantly different between silodosin and tamsulosin (risk ratio: 1.07, 95% confidence interval [CI]: 0.91–1.26; moderate SOE). Silodosin and tamsulosin (all dose levels) were not significantly different in improving mean I-PSS scores (weighted mean difference [WMD]: –0.52, 95% CI: –1.58 to 0.54; moderate SOE). Results were similar in analyses restricted to trials using tamsulosin 0.4 mg.

Overall withdrawals (for any reason) and rates of participants with ≥1 AEs were not significantly different between treatments (low and moderate SOE, respectively). Withdrawals due to AEs were more frequent for silodosin versus tamsulosin (risk ratio: 1.96, 95% CI: 1.04–3.71; moderate SOE). The most common AE was abnormal ejaculation, reported in 16% of patients on silodosin versus 2% of patients on tamsulosin.

3.3 Darifenacin plus ABs versus ABs alone

Two 146 12-wk trials ( n = 161) compared darifenacin/AB combination 147 therapy with AB monotherapy in men with LUTS and overactive bladder (OAB) symptoms attributed to BPH [23 24] . Participants with a baseline postvoid residual of >150 ml were excluded. RoB was low in one trial [24] and moderate in the other [23] . SOE was judged insufficient for all outcomes.

3.4 Fesoterodine plus ABs versus ABs alone

Two trials ( n = 990) compared fesoterodine/AB combination therapy with AB monotherapy ( Table 1 ) in men with LUTS and OAB symptoms [25 26] . Overall RoB was moderate for one trial [25] and high for the other [26] . Improvement in mean I-PSS scores was similar with fesoterodine/AB combination and AB monotherapy (low SOE; Table 3 ). The mean difference in the large moderate RoB trial was 0.0 (95% CI: –0.83 to 0.83) and –1.70 (95% CI: –5.85 to 2.46) for the small high RoB trial. Overall withdrawals, withdrawals due to AEs, and the number of participants with ≥1 AE were more frequent in the fesoterodine arm; SOE was judged as low for all three outcomes.

3.5 Oxybutynin plus ABs versus ABs alone

One 12-wk trial ( n = 420) trial compared oxybutynin 10 mg tablets plus tamsulosin 0.4 mg with tamsulosin 0.4 mg alone [27] . Individuals with a baseline postvoid residual of >200 ml were excluded. RoB was moderate. SOE was judged insufficient for all outcomes.

3.6 Solifenacin plus ABs versus ABs alone

Seven 12-wk trials 28 29 30 31 32 33 34 randomized men with LUTS and OAB symptoms ( n = 3147) to solifenacin plus tamsulosin versus tamsulosin monotherapy ( Table 1 ). Five trials examined solifenacin 5 mg 28 29 30 31 34] and two examined solifenacin 6 mg [32 33] . The overall RoB was moderate.

Combination therapy was similar to AB monotherapy in improving LUTS (moderate SOE; Table 4 ). Improvement in mean I-PSS score from baseline was similar with solifenacin 5 mg or 6 mg plus tamsulosin 0.2 mg or 0.4 mg versus tamsulosin alone (WMD: –0.29, 95% CI: –0.88 to 0.30; moderate SOE). Combination therapy lowered I-PSS QoL score more than tamsulosin, but the difference between groups was not clinically significant based on predetermined thresholds (moderate SOE). Evidence from four trials using solifenacin 3–9 mg [29 32 33 34 reported no statistical difference between treatment groups in rates of AUR, but evidence was judged insufficient to draw conclusions due to imprecision.

Withdrawal for any reason or due to AEs was similar with both treatments (low SOE). More participants reported one or more AEs with combination treatment than monotherapy (moderate SOE).

3.7 Tolterodine plus ABs versus ABs alone

Four trials randomized men with LUTS and OAB symptoms ( n = 1338) to a combination of tolterodine 4 mg plus AB versus AB monotherapy with tamsulosin, doxazosin, or alfuzosin ( Table 5 ) 35 36 37 38 . Overall RoB for three trials was low 35 36 37 and one trial had a high RoB [38] .

Responder analysis was provided in only the high RoB trial [38] with a response defined as a 3-point improvement in I-PSS score from baseline ( Table 3 ). The proportion of responders was greater in the combination group than the AB monotherapy group (77% vs 29%), but the SOE for the comparison was judged as insufficient for this outcome due to a high RoB and unknown consistency. Mean changes in I-PSS scores were similar between treatment groups (WMD: –0.19, 95% CI: –1.08 to 0.68; moderate SOE). The mean change in I-PSS QoL was also not different between treatment groups (WMD: –0.34, 95% CI: –1.14 to 0.46; low SOE) 35 36 37 .

There were six incidences of AUR in the combination group versus two in the monotherapy group, which was judged as insufficient evidence 35 36 37 . Withdrawal for any reason (low SOE), withdrawal due to adverse effects (low SOE), and rates of ≥1 AEs (insufficient SOE) were not different between groups [35] .

3.8 Trospium plus ABs versus ABs alone

One 12-wk trial ( n = 58) compared trospium 45 mg daily doses with AB to AB monotherapy in men with LUTS and OAB symptoms attributed to BPH [39] . Individuals with a baseline postvoid residual of >100 ml were excluded. RoB was moderate.

Evidence was insufficient to assess efficacy for any outcome. One or more AE were reported in nine (35%) trospium participants versus five (23%) placebo patients.

3.9 Mirabegron plus AB versus AB alone

We found one 8-wk trial [40] ( n = 94), which compared 50 mg of mirabegron combined with 0.2 mg tamsulosin versus tamsulosin monotherapy in males with LUTS and OAB symptoms attributed to BPH. Patients with postvoid residual urine volume >100 ml were excluded. The study was judged to be at high RoB. The SOE was judged insufficient for all predetermined patient-important outcomes. We found no comparative trials combining mirabegron with other ABs for this indication.

3.10 Tadalafil versus tamsulosin

Four 3-mo trials compared tadalafil 2.5 mg, 5 mg, or 10 mg daily with tamsulosin 0.2 mg or 0.4 mg daily ( Table 1 ) 41 42 43 44 . Most participants had a history of erectile dysfunction (ED) [41 43 44] . The most frequently investigated dose level of tadalafil was 5 mg; one trial included a 2.5-mg dose level [42] and one trial evaluated 10 mg [41] . Overall RoB ranged from low to high for the four trials.

Tadalafil 5 mg and tamsulosin were similar in improving mean I-PSS scores (WMD: –0.07, 95% CI: –2.12 to 2.23; moderate SOE) and I-PSS QoL (WMD: –0.01, 95% CI: –0.75 to 0.73; low SOE). Evidence was insufficient for the outcomes of study withdrawal for any reason and the proportion of participants reporting at least one AE, but withdrawal due to AEs was higher with tadalafil (3% vs 1%; moderate SOE; Table 6 ).

3.11 Tadalafil versus alfuzosin

Two 3-mo trials ( n = 93) compared tadalafil with alfuzosin 10 mg daily ( Table 1 ) [45 46] . Kumar et al [45] compared tadalafil 10 mg daily with alfuzosin 10 mg daily. Liguori et al [46] compared tadalafil 20 mg taken on alternate days with alfuzosin 10 mg daily. All participants had a history of ED. Both trials were open label with a high overall RoB.

Alfuzosin 10 mg improved mean I-PSS scores more than tadalafil 10 mg or 20 mg (low SOE; Table 7 ). Mean reductions in I-PSS scores were 4.1 and 7.2 points with tadalafil and alfuzosin, respectively, favoring alfuzosin. I-PSS QoL also improved more with alfuzosin than tadalafil (low SOE). Study withdrawal for any reason and withdrawal due to an AE were no different with tadalafil or alfuzosin (insufficient SOE).

3.12 Sildenafil versus ABs

Three trials ( n = 181) compared sildenafil versus an AB 47 48 49 . One compared sildenafil 25 mg daily with alfuzosin 10 mg daily over 12 wk [48] and one compared sildenafil 25 mg taken 4 d/wk with tamsulosin 0.4 mg daily over 8 wk [49] . Abolyosr et al [47] compared sildenafil 50 mg to a low dose of doxazosin 2 mg over 16 wk; frequency of administration was not reported. All participants had a history of ED. All trials were open label and the overall RoB was high.

Mean change in I-PSS scores was –2.2 with sildenafil and –3.2 with alfuzosin or doxazosin (insufficient SOE) [47 48] . Mean change in I-PSS scores with sildenafil 25 mg was 4.0 points versus 5.4 points for tamsulosin 0.4 mg (insufficient SOE) [49] .

Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.13 Tadalafil AB combination versus AB monotherapy

Four trials ( n = 224) compared tadalafil combined with an AB versus AB monotherapy [41 45 46 50] . Two 3-mo trials compared tadalafil 10 mg daily [45] or 20 mg on alternate days [46] combined with alfuzosin 10 mg to alfuzosin 10 mg monotherapy. Two trials evaluated tadalafil combined with tamsulosin 0.4 mg versus tamsulosin 0.4 mg monotherapy: a 1-mo trial evaluated tadalafil 5 mg daily [50] and a 4-mo trial evaluated tadalafil 10 mg daily [41] . Nearly all participants had a history of ED [41 45 46] . All trials were open label except Regadas et al [50] and the overall RoB therefore ranged from moderate to high.

Results with combination therapy were similar to AB monotherapy (tadalafil 5–20 mg combined with AB was no different than AB monotherapy in improving mean I-PSS scores from baseline [WMD: –2.0, 95% CI: –4.03 to –0.00; insufficient SOE]). Mean reductions in I-PSS scores were similar; 10.4 and 8.6 with combination and monotherapy. Improvement in mean I-PSS QoL scores was also not significantly different between combination treatment and monotherapy; however, only open label (high RoB) trials reported this outcome (low SOE) [41 45 46] . Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.14 Sildenafil AB combination versus AB monotherapy

Four trials ( n = 281) compared sildenafil combined with an AB versus AB monotherapy 47 48 49 51] . Two 3-mo trials evaluated sildenafil combined with alfuzosin 10 mg versus alfuzosin monotherapy; one used daily sildenafil 25 mg [48] and the other used sildenafil 50 mg (dosing frequency not reported) [51] . One 4-mo trial evaluated sildenafil 50 mg combined with doxazosin 2 mg versus doxazosin alone, but dosing frequency was not reported [47] . An 8-wk trial evaluated sildenafil 25 mg taken 4 d/wk combined with tamsulosin 0.4 mg daily compared with tamsulosin monotherapy [49] . Three trials enrolled men with a history of ED 47 48 49 . All trials were open label or otherwise inadequately blinded and enrolled patients after they failed to respond to AB monotherapy. Overall RoB was mostly high.

Mean reductions in I-PSS scores were 5.4 with combination and 3.9 with monotherapy, with both treatments exceeding MDD. However, the SOE to assess the comparative effectiveness was insufficient due to the study limitations and imprecision in measurement [47 48 51] . In the 8-wk trial without data sufficient for pooling, improvement in mean I-PSS scores was similar with combination or monotherapy (–6.4 vs –5.4) [49] . Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.15 Vardenafil AB combination versus AB monotherapy

One double-blinded trial ( n = 60) compared vardenafil 10 mg daily combined with tamsulosin 0.4 mg to tamsulosin monotherapy over 12 wk [52] . The overall RoB was moderate. Mean reductions in I-PSS scores were 5.8 with combination treatment and 3.7 with monotherapy, both achieving MDD (insufficient SOE).

One withdrawal was reported with tamsulosin. No participant withdrew due to AEs. Persistent AEs were reported in three participants with combination therapy (headache with flushing, headache with stomach pain, and stomach pain) and two with tamsulosin (headache and flushing). No serious AEs were reported.

3.16 Tadalafil plus 5-ARI combination versus 5-ARI monotherapy and tadalafil versus 5-ARI/AB combination

The evidence from a single trial of tadalafil 5-ARI combination versus 5-ARI monotherapy [53] as well as that from a single trial comparing tadalafil versus 5-ARI/AB combination [54] was insufficient for both comparative effectiveness and safety.

3.17 Longer-term harms from observational studies

We identified two observational studies reporting longer-term AEs related to silodosin treatment [55 56] . In a 40-wk open label extension (cumulative treatment duration of 52 wk) of a previous RCT, 435 patients completed the extension in which a total of 431 experienced 924 AEs [55] . Twenty-nine patients (4.4%) experienced serious AEs including two deaths; none of the serious AEs were considered drug related, although no criteria were reported. The second study reviewed FDA data for AEs associated with ABs and found the evidence of silodosin insuffient to compare with other ABs [56] .

We identified one study examining long-term AEs associated with solifenacin/AB combination therapy, reporting an open extension from a previous RCT for a subset of patients with inclusion criteria that included a postvoid residual ≤150 ml [57] . Forty-seven percent of participants who continued solifenacin/AB combination treatment reported treatment-emergent AEs, most commonly dry mouth, constipation, and dyspepsia. In addition, 86 serious AEs occurred in 64 patients and included three deaths, six cases of AUR (0.7%), and three cases of intervertebral disc protrusion.

We found two longer-term observational studies on tadalafil [58 59] . In a 42-wk open label extension of a previous trial, 59% of 394 participants reported at least one AE and 9% withdrew due to an AE. Serious AEs were reported in 3% (11 participants) [58] . In another open extension study in a subset of 229 of 886 original participants, nearly 5% experienced serious AEs [59] .

2.1 Protocol

We developed an a priori written protocol, together with a technical report that incorporated input from key stakeholders, a multidisciplinary Technical Expert Panel, and public comment (available at the Agency for Healthcare Research and Quality [AHRQ] website http://effectivehealthcare.ahrq.gov/index.cfm/search-for-guides-reviews-and-reports/?productid=2067&pageaction=displayproduct ).

2.2 Eligibility criteria

Based on our Population, Interventions, Comparisons, Outcomes, Timing, and Setting criteria (Supplementary data) we included randomized controlled trials (RCTs) that tested comparative effectiveness of treatments involving newer drugs in men aged ≥45 yr with LUTS attributed to BPH. We defined these newer drugs as those that were Food and Drug Administration (FDA) approved for BPH since 2008 or which, though not FDA approved for BPH, have been studied for the treatment of BPH since 2008 and were selected through a formal process of stakeholder involvement (Supplementary data). Comparators included medications FDA approved for BPH before 2008. Included RCTs were at least 1 mo in duration with no minimum sample size. We additionally searched for large ( n ≥ 100 patients), longer-term (≥ 1 yr duration) observational studies to assess long-term or rare treatment associated harms only. We limited inclusion to English language articles.

The primary predefined outcomes of interest were changes reflecting clinically important differences (Supplementary data) in validated measures to assess LUTS (International Prostate Symptom Score [I-PSS]: score ranges 0–35 with higher scores indicating more severe symptoms; or American Urological Association Symptom Index scores), prostate-related bother or quality of life (QoL; I-PSS QoL question; BPH/LUTS impact scale), as well as rates of disease progression and/or treatment failure (prevention/delay of need for surgical intervention; acute urinary retention [AUR]). We also assessed common and serious medication adverse effects (AEs).

2.3 Information sources and literature search

We searched Ovid Medline, Ovid Embase, and the Cochrane Central Register of Controlled Trials with filters for study design (Supplementary data), to identify relevant RCTs published through June 20, 2016. We also searched for relevant systematic reviews and other key references. Lastly, we searched the Clinical Trials ( www.clinicaltrials.gov ) and the FDA ( www.fda.gov/Drugs ) websites to identify additional completed and ongoing studies.

2.4 Study selection process, data extraction, and risk of bias in studies

Two independent investigators screened titles and abstracts to identify studies meeting the eligibility criteria. Data were extracted by one investigator and reviewed and verified for accuracy by a second investigator. Risk of bias (RoB) of eligible studies was assessed using AHRQ guidance by one investigator and reviewed by a second [7] .

2.5 Synthesis of results

We assessed clinical and methodological heterogeneity and variation in effect size to determine the appropriateness of pooling data [8] . When three or more trials reported similar comparisons and outcomes, data were pooled using a Hartung, Knapp, Sidik, and Jonkman method [9] random effects model for proportion of I-PSS responders or mean changes in I-PSS scores in Stata (StataCorp., College Station, TX, USA). We pooled other outcomes in RevMan (RevMan, Spartanburg, SC, USA) [10] and converted DerSimonian-Laird random effects confidence intervals to Hartung, Knapp, Sidik, and Jonkman confidence intervals using an excel spreadsheet provided in Inthout et al [9] . We assessed between study variance with Tau ² and measured the magnitude of heterogeneity with the I ² statistic. If substantial heterogeneity was present (ie, I ² ≥ 70%), we stratified the results to assess treatment effects based on patient or study characteristics and/or explored sensitivity analyses [8 11] . We pooled across different ABs unless there were at least three trials for a given agent. We interpreted efficacy and comparative effectiveness using established thresholds indicating clinical significance (Supplementary data).

For the body of evidence from RCTs, we rated our confidence in the estimates of effect for the primary outcomes as overall strength of evidence (SOE) as high, moderate, low, or insufficient (Supplementary data) [12] . For observational studies, we did not formally assess SOE, but provided descriptive information in narrative form.

3.1 Search results

Our literature search identified 1139 references, of which 124 were selected for full-text review ( Fig. 1 ). This process mapped to 43 unique RCTs. In addition, we identified five longer duration (≥ 1 yr) observational studies reporting AEs.

 

Fig. 1
Figure 1: Literature flow diagram. The results are presented separately for each of four drug classes (new alpha-blockers, anticholinergics, beta-3 agonists, and phosphodiesterase type-5 inhibitor), and specific drugs are listed within each class. The outcomes addressed by the three key questions are discussed within each drug-specific section.

BPH = benign prostatic hyperplasia; RCT = randomized controlled trial.

3.2 Silodosin versus tamsulosin

Ten trials randomized men with LUTS attributed to BPH ( n = 1799) to silodosin 8 mg daily versus tamsulosin 0.2–0.4 mg daily 13 14 15 16 17 18 19 20 21 22 . Table 1 provides baseline characteristics. Overall, the RoB was low in two trials [13 18] , moderate in six trials 14 15 16 20 21 22 , and high in two trials [17 19] .

Three trials conducted responder analyses (defined as > 25% reduction in I-PSS score; Table 2 ) [16 18 20] . Response to treatment was not significantly different between silodosin and tamsulosin (risk ratio: 1.07, 95% confidence interval [CI]: 0.91–1.26; moderate SOE). Silodosin and tamsulosin (all dose levels) were not significantly different in improving mean I-PSS scores (weighted mean difference [WMD]: –0.52, 95% CI: –1.58 to 0.54; moderate SOE). Results were similar in analyses restricted to trials using tamsulosin 0.4 mg.

Overall withdrawals (for any reason) and rates of participants with ≥1 AEs were not significantly different between treatments (low and moderate SOE, respectively). Withdrawals due to AEs were more frequent for silodosin versus tamsulosin (risk ratio: 1.96, 95% CI: 1.04–3.71; moderate SOE). The most common AE was abnormal ejaculation, reported in 16% of patients on silodosin versus 2% of patients on tamsulosin.

3.3 Darifenacin plus ABs versus ABs alone

Two 146 12-wk trials ( n = 161) compared darifenacin/AB combination 147 therapy with AB monotherapy in men with LUTS and overactive bladder (OAB) symptoms attributed to BPH [23 24] . Participants with a baseline postvoid residual of >150 ml were excluded. RoB was low in one trial [24] and moderate in the other [23] . SOE was judged insufficient for all outcomes.

3.4 Fesoterodine plus ABs versus ABs alone

Two trials ( n = 990) compared fesoterodine/AB combination therapy with AB monotherapy ( Table 1 ) in men with LUTS and OAB symptoms [25 26] . Overall RoB was moderate for one trial [25] and high for the other [26] . Improvement in mean I-PSS scores was similar with fesoterodine/AB combination and AB monotherapy (low SOE; Table 3 ). The mean difference in the large moderate RoB trial was 0.0 (95% CI: –0.83 to 0.83) and –1.70 (95% CI: –5.85 to 2.46) for the small high RoB trial. Overall withdrawals, withdrawals due to AEs, and the number of participants with ≥1 AE were more frequent in the fesoterodine arm; SOE was judged as low for all three outcomes.

3.5 Oxybutynin plus ABs versus ABs alone

One 12-wk trial ( n = 420) trial compared oxybutynin 10 mg tablets plus tamsulosin 0.4 mg with tamsulosin 0.4 mg alone [27] . Individuals with a baseline postvoid residual of >200 ml were excluded. RoB was moderate. SOE was judged insufficient for all outcomes.

3.6 Solifenacin plus ABs versus ABs alone

Seven 12-wk trials 28 29 30 31 32 33 34 randomized men with LUTS and OAB symptoms ( n = 3147) to solifenacin plus tamsulosin versus tamsulosin monotherapy ( Table 1 ). Five trials examined solifenacin 5 mg 28 29 30 31 34] and two examined solifenacin 6 mg [32 33] . The overall RoB was moderate.

Combination therapy was similar to AB monotherapy in improving LUTS (moderate SOE; Table 4 ). Improvement in mean I-PSS score from baseline was similar with solifenacin 5 mg or 6 mg plus tamsulosin 0.2 mg or 0.4 mg versus tamsulosin alone (WMD: –0.29, 95% CI: –0.88 to 0.30; moderate SOE). Combination therapy lowered I-PSS QoL score more than tamsulosin, but the difference between groups was not clinically significant based on predetermined thresholds (moderate SOE). Evidence from four trials using solifenacin 3–9 mg [29 32 33 34 reported no statistical difference between treatment groups in rates of AUR, but evidence was judged insufficient to draw conclusions due to imprecision.

Withdrawal for any reason or due to AEs was similar with both treatments (low SOE). More participants reported one or more AEs with combination treatment than monotherapy (moderate SOE).

3.7 Tolterodine plus ABs versus ABs alone

Four trials randomized men with LUTS and OAB symptoms ( n = 1338) to a combination of tolterodine 4 mg plus AB versus AB monotherapy with tamsulosin, doxazosin, or alfuzosin ( Table 5 ) 35 36 37 38 . Overall RoB for three trials was low 35 36 37 and one trial had a high RoB [38] .

Responder analysis was provided in only the high RoB trial [38] with a response defined as a 3-point improvement in I-PSS score from baseline ( Table 3 ). The proportion of responders was greater in the combination group than the AB monotherapy group (77% vs 29%), but the SOE for the comparison was judged as insufficient for this outcome due to a high RoB and unknown consistency. Mean changes in I-PSS scores were similar between treatment groups (WMD: –0.19, 95% CI: –1.08 to 0.68; moderate SOE). The mean change in I-PSS QoL was also not different between treatment groups (WMD: –0.34, 95% CI: –1.14 to 0.46; low SOE) 35 36 37 .

There were six incidences of AUR in the combination group versus two in the monotherapy group, which was judged as insufficient evidence 35 36 37 . Withdrawal for any reason (low SOE), withdrawal due to adverse effects (low SOE), and rates of ≥1 AEs (insufficient SOE) were not different between groups [35] .

3.8 Trospium plus ABs versus ABs alone

One 12-wk trial ( n = 58) compared trospium 45 mg daily doses with AB to AB monotherapy in men with LUTS and OAB symptoms attributed to BPH [39] . Individuals with a baseline postvoid residual of >100 ml were excluded. RoB was moderate.

Evidence was insufficient to assess efficacy for any outcome. One or more AE were reported in nine (35%) trospium participants versus five (23%) placebo patients.

3.9 Mirabegron plus AB versus AB alone

We found one 8-wk trial [40] ( n = 94), which compared 50 mg of mirabegron combined with 0.2 mg tamsulosin versus tamsulosin monotherapy in males with LUTS and OAB symptoms attributed to BPH. Patients with postvoid residual urine volume >100 ml were excluded. The study was judged to be at high RoB. The SOE was judged insufficient for all predetermined patient-important outcomes. We found no comparative trials combining mirabegron with other ABs for this indication.

3.10 Tadalafil versus tamsulosin

Four 3-mo trials compared tadalafil 2.5 mg, 5 mg, or 10 mg daily with tamsulosin 0.2 mg or 0.4 mg daily ( Table 1 ) 41 42 43 44 . Most participants had a history of erectile dysfunction (ED) [41 43 44] . The most frequently investigated dose level of tadalafil was 5 mg; one trial included a 2.5-mg dose level [42] and one trial evaluated 10 mg [41] . Overall RoB ranged from low to high for the four trials.

Tadalafil 5 mg and tamsulosin were similar in improving mean I-PSS scores (WMD: –0.07, 95% CI: –2.12 to 2.23; moderate SOE) and I-PSS QoL (WMD: –0.01, 95% CI: –0.75 to 0.73; low SOE). Evidence was insufficient for the outcomes of study withdrawal for any reason and the proportion of participants reporting at least one AE, but withdrawal due to AEs was higher with tadalafil (3% vs 1%; moderate SOE; Table 6 ).

3.11 Tadalafil versus alfuzosin

Two 3-mo trials ( n = 93) compared tadalafil with alfuzosin 10 mg daily ( Table 1 ) [45 46] . Kumar et al [45] compared tadalafil 10 mg daily with alfuzosin 10 mg daily. Liguori et al [46] compared tadalafil 20 mg taken on alternate days with alfuzosin 10 mg daily. All participants had a history of ED. Both trials were open label with a high overall RoB.

Alfuzosin 10 mg improved mean I-PSS scores more than tadalafil 10 mg or 20 mg (low SOE; Table 7 ). Mean reductions in I-PSS scores were 4.1 and 7.2 points with tadalafil and alfuzosin, respectively, favoring alfuzosin. I-PSS QoL also improved more with alfuzosin than tadalafil (low SOE). Study withdrawal for any reason and withdrawal due to an AE were no different with tadalafil or alfuzosin (insufficient SOE).

3.12 Sildenafil versus ABs

Three trials ( n = 181) compared sildenafil versus an AB 47 48 49 . One compared sildenafil 25 mg daily with alfuzosin 10 mg daily over 12 wk [48] and one compared sildenafil 25 mg taken 4 d/wk with tamsulosin 0.4 mg daily over 8 wk [49] . Abolyosr et al [47] compared sildenafil 50 mg to a low dose of doxazosin 2 mg over 16 wk; frequency of administration was not reported. All participants had a history of ED. All trials were open label and the overall RoB was high.

Mean change in I-PSS scores was –2.2 with sildenafil and –3.2 with alfuzosin or doxazosin (insufficient SOE) [47 48] . Mean change in I-PSS scores with sildenafil 25 mg was 4.0 points versus 5.4 points for tamsulosin 0.4 mg (insufficient SOE) [49] .

Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.13 Tadalafil AB combination versus AB monotherapy

Four trials ( n = 224) compared tadalafil combined with an AB versus AB monotherapy [41 45 46 50] . Two 3-mo trials compared tadalafil 10 mg daily [45] or 20 mg on alternate days [46] combined with alfuzosin 10 mg to alfuzosin 10 mg monotherapy. Two trials evaluated tadalafil combined with tamsulosin 0.4 mg versus tamsulosin 0.4 mg monotherapy: a 1-mo trial evaluated tadalafil 5 mg daily [50] and a 4-mo trial evaluated tadalafil 10 mg daily [41] . Nearly all participants had a history of ED [41 45 46] . All trials were open label except Regadas et al [50] and the overall RoB therefore ranged from moderate to high.

Results with combination therapy were similar to AB monotherapy (tadalafil 5–20 mg combined with AB was no different than AB monotherapy in improving mean I-PSS scores from baseline [WMD: –2.0, 95% CI: –4.03 to –0.00; insufficient SOE]). Mean reductions in I-PSS scores were similar; 10.4 and 8.6 with combination and monotherapy. Improvement in mean I-PSS QoL scores was also not significantly different between combination treatment and monotherapy; however, only open label (high RoB) trials reported this outcome (low SOE) [41 45 46] . Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.14 Sildenafil AB combination versus AB monotherapy

Four trials ( n = 281) compared sildenafil combined with an AB versus AB monotherapy 47 48 49 51] . Two 3-mo trials evaluated sildenafil combined with alfuzosin 10 mg versus alfuzosin monotherapy; one used daily sildenafil 25 mg [48] and the other used sildenafil 50 mg (dosing frequency not reported) [51] . One 4-mo trial evaluated sildenafil 50 mg combined with doxazosin 2 mg versus doxazosin alone, but dosing frequency was not reported [47] . An 8-wk trial evaluated sildenafil 25 mg taken 4 d/wk combined with tamsulosin 0.4 mg daily compared with tamsulosin monotherapy [49] . Three trials enrolled men with a history of ED 47 48 49 . All trials were open label or otherwise inadequately blinded and enrolled patients after they failed to respond to AB monotherapy. Overall RoB was mostly high.

Mean reductions in I-PSS scores were 5.4 with combination and 3.9 with monotherapy, with both treatments exceeding MDD. However, the SOE to assess the comparative effectiveness was insufficient due to the study limitations and imprecision in measurement [47 48 51] . In the 8-wk trial without data sufficient for pooling, improvement in mean I-PSS scores was similar with combination or monotherapy (–6.4 vs –5.4) [49] . Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.15 Vardenafil AB combination versus AB monotherapy

One double-blinded trial ( n = 60) compared vardenafil 10 mg daily combined with tamsulosin 0.4 mg to tamsulosin monotherapy over 12 wk [52] . The overall RoB was moderate. Mean reductions in I-PSS scores were 5.8 with combination treatment and 3.7 with monotherapy, both achieving MDD (insufficient SOE).

One withdrawal was reported with tamsulosin. No participant withdrew due to AEs. Persistent AEs were reported in three participants with combination therapy (headache with flushing, headache with stomach pain, and stomach pain) and two with tamsulosin (headache and flushing). No serious AEs were reported.

3.16 Tadalafil plus 5-ARI combination versus 5-ARI monotherapy and tadalafil versus 5-ARI/AB combination

The evidence from a single trial of tadalafil 5-ARI combination versus 5-ARI monotherapy [53] as well as that from a single trial comparing tadalafil versus 5-ARI/AB combination [54] was insufficient for both comparative effectiveness and safety.

3.17 Longer-term harms from observational studies

We identified two observational studies reporting longer-term AEs related to silodosin treatment [55 56] . In a 40-wk open label extension (cumulative treatment duration of 52 wk) of a previous RCT, 435 patients completed the extension in which a total of 431 experienced 924 AEs [55] . Twenty-nine patients (4.4%) experienced serious AEs including two deaths; none of the serious AEs were considered drug related, although no criteria were reported. The second study reviewed FDA data for AEs associated with ABs and found the evidence of silodosin insuffient to compare with other ABs [56] .

We identified one study examining long-term AEs associated with solifenacin/AB combination therapy, reporting an open extension from a previous RCT for a subset of patients with inclusion criteria that included a postvoid residual ≤150 ml [57] . Forty-seven percent of participants who continued solifenacin/AB combination treatment reported treatment-emergent AEs, most commonly dry mouth, constipation, and dyspepsia. In addition, 86 serious AEs occurred in 64 patients and included three deaths, six cases of AUR (0.7%), and three cases of intervertebral disc protrusion.

We found two longer-term observational studies on tadalafil [58 59] . In a 42-wk open label extension of a previous trial, 59% of 394 participants reported at least one AE and 9% withdrew due to an AE. Serious AEs were reported in 3% (11 participants) [58] . In another open extension study in a subset of 229 of 886 original participants, nearly 5% experienced serious AEs [59] .

2.1 Protocol

We developed an a priori written protocol, together with a technical report that incorporated input from key stakeholders, a multidisciplinary Technical Expert Panel, and public comment (available at the Agency for Healthcare Research and Quality [AHRQ] website http://effectivehealthcare.ahrq.gov/index.cfm/search-for-guides-reviews-and-reports/?productid=2067&pageaction=displayproduct ).

2.2 Eligibility criteria

Based on our Population, Interventions, Comparisons, Outcomes, Timing, and Setting criteria (Supplementary data) we included randomized controlled trials (RCTs) that tested comparative effectiveness of treatments involving newer drugs in men aged ≥45 yr with LUTS attributed to BPH. We defined these newer drugs as those that were Food and Drug Administration (FDA) approved for BPH since 2008 or which, though not FDA approved for BPH, have been studied for the treatment of BPH since 2008 and were selected through a formal process of stakeholder involvement (Supplementary data). Comparators included medications FDA approved for BPH before 2008. Included RCTs were at least 1 mo in duration with no minimum sample size. We additionally searched for large ( n ≥ 100 patients), longer-term (≥ 1 yr duration) observational studies to assess long-term or rare treatment associated harms only. We limited inclusion to English language articles.

The primary predefined outcomes of interest were changes reflecting clinically important differences (Supplementary data) in validated measures to assess LUTS (International Prostate Symptom Score [I-PSS]: score ranges 0–35 with higher scores indicating more severe symptoms; or American Urological Association Symptom Index scores), prostate-related bother or quality of life (QoL; I-PSS QoL question; BPH/LUTS impact scale), as well as rates of disease progression and/or treatment failure (prevention/delay of need for surgical intervention; acute urinary retention [AUR]). We also assessed common and serious medication adverse effects (AEs).

2.3 Information sources and literature search

We searched Ovid Medline, Ovid Embase, and the Cochrane Central Register of Controlled Trials with filters for study design (Supplementary data), to identify relevant RCTs published through June 20, 2016. We also searched for relevant systematic reviews and other key references. Lastly, we searched the Clinical Trials ( www.clinicaltrials.gov ) and the FDA ( www.fda.gov/Drugs ) websites to identify additional completed and ongoing studies.

2.4 Study selection process, data extraction, and risk of bias in studies

Two independent investigators screened titles and abstracts to identify studies meeting the eligibility criteria. Data were extracted by one investigator and reviewed and verified for accuracy by a second investigator. Risk of bias (RoB) of eligible studies was assessed using AHRQ guidance by one investigator and reviewed by a second [7] .

2.5 Synthesis of results

We assessed clinical and methodological heterogeneity and variation in effect size to determine the appropriateness of pooling data [8] . When three or more trials reported similar comparisons and outcomes, data were pooled using a Hartung, Knapp, Sidik, and Jonkman method [9] random effects model for proportion of I-PSS responders or mean changes in I-PSS scores in Stata (StataCorp., College Station, TX, USA). We pooled other outcomes in RevMan (RevMan, Spartanburg, SC, USA) [10] and converted DerSimonian-Laird random effects confidence intervals to Hartung, Knapp, Sidik, and Jonkman confidence intervals using an excel spreadsheet provided in Inthout et al [9] . We assessed between study variance with Tau ² and measured the magnitude of heterogeneity with the I ² statistic. If substantial heterogeneity was present (ie, I ² ≥ 70%), we stratified the results to assess treatment effects based on patient or study characteristics and/or explored sensitivity analyses [8 11] . We pooled across different ABs unless there were at least three trials for a given agent. We interpreted efficacy and comparative effectiveness using established thresholds indicating clinical significance (Supplementary data).

For the body of evidence from RCTs, we rated our confidence in the estimates of effect for the primary outcomes as overall strength of evidence (SOE) as high, moderate, low, or insufficient (Supplementary data) [12] . For observational studies, we did not formally assess SOE, but provided descriptive information in narrative form.

3.1 Search results

Our literature search identified 1139 references, of which 124 were selected for full-text review ( Fig. 1 ). This process mapped to 43 unique RCTs. In addition, we identified five longer duration (≥ 1 yr) observational studies reporting AEs.

 

Fig. 1
Figure 1: Literature flow diagram. The results are presented separately for each of four drug classes (new alpha-blockers, anticholinergics, beta-3 agonists, and phosphodiesterase type-5 inhibitor), and specific drugs are listed within each class. The outcomes addressed by the three key questions are discussed within each drug-specific section.

BPH = benign prostatic hyperplasia; RCT = randomized controlled trial.

3.2 Silodosin versus tamsulosin

Ten trials randomized men with LUTS attributed to BPH ( n = 1799) to silodosin 8 mg daily versus tamsulosin 0.2–0.4 mg daily 13 14 15 16 17 18 19 20 21 22 . Table 1 provides baseline characteristics. Overall, the RoB was low in two trials [13 18] , moderate in six trials 14 15 16 20 21 22 , and high in two trials [17 19] .

Three trials conducted responder analyses (defined as > 25% reduction in I-PSS score; Table 2 ) [16 18 20] . Response to treatment was not significantly different between silodosin and tamsulosin (risk ratio: 1.07, 95% confidence interval [CI]: 0.91–1.26; moderate SOE). Silodosin and tamsulosin (all dose levels) were not significantly different in improving mean I-PSS scores (weighted mean difference [WMD]: –0.52, 95% CI: –1.58 to 0.54; moderate SOE). Results were similar in analyses restricted to trials using tamsulosin 0.4 mg.

Overall withdrawals (for any reason) and rates of participants with ≥1 AEs were not significantly different between treatments (low and moderate SOE, respectively). Withdrawals due to AEs were more frequent for silodosin versus tamsulosin (risk ratio: 1.96, 95% CI: 1.04–3.71; moderate SOE). The most common AE was abnormal ejaculation, reported in 16% of patients on silodosin versus 2% of patients on tamsulosin.

3.3 Darifenacin plus ABs versus ABs alone

Two 146 12-wk trials ( n = 161) compared darifenacin/AB combination 147 therapy with AB monotherapy in men with LUTS and overactive bladder (OAB) symptoms attributed to BPH [23 24] . Participants with a baseline postvoid residual of >150 ml were excluded. RoB was low in one trial [24] and moderate in the other [23] . SOE was judged insufficient for all outcomes.

3.4 Fesoterodine plus ABs versus ABs alone

Two trials ( n = 990) compared fesoterodine/AB combination therapy with AB monotherapy ( Table 1 ) in men with LUTS and OAB symptoms [25 26] . Overall RoB was moderate for one trial [25] and high for the other [26] . Improvement in mean I-PSS scores was similar with fesoterodine/AB combination and AB monotherapy (low SOE; Table 3 ). The mean difference in the large moderate RoB trial was 0.0 (95% CI: –0.83 to 0.83) and –1.70 (95% CI: –5.85 to 2.46) for the small high RoB trial. Overall withdrawals, withdrawals due to AEs, and the number of participants with ≥1 AE were more frequent in the fesoterodine arm; SOE was judged as low for all three outcomes.

3.5 Oxybutynin plus ABs versus ABs alone

One 12-wk trial ( n = 420) trial compared oxybutynin 10 mg tablets plus tamsulosin 0.4 mg with tamsulosin 0.4 mg alone [27] . Individuals with a baseline postvoid residual of >200 ml were excluded. RoB was moderate. SOE was judged insufficient for all outcomes.

3.6 Solifenacin plus ABs versus ABs alone

Seven 12-wk trials 28 29 30 31 32 33 34 randomized men with LUTS and OAB symptoms ( n = 3147) to solifenacin plus tamsulosin versus tamsulosin monotherapy ( Table 1 ). Five trials examined solifenacin 5 mg 28 29 30 31 34] and two examined solifenacin 6 mg [32 33] . The overall RoB was moderate.

Combination therapy was similar to AB monotherapy in improving LUTS (moderate SOE; Table 4 ). Improvement in mean I-PSS score from baseline was similar with solifenacin 5 mg or 6 mg plus tamsulosin 0.2 mg or 0.4 mg versus tamsulosin alone (WMD: –0.29, 95% CI: –0.88 to 0.30; moderate SOE). Combination therapy lowered I-PSS QoL score more than tamsulosin, but the difference between groups was not clinically significant based on predetermined thresholds (moderate SOE). Evidence from four trials using solifenacin 3–9 mg [29 32 33 34 reported no statistical difference between treatment groups in rates of AUR, but evidence was judged insufficient to draw conclusions due to imprecision.

Withdrawal for any reason or due to AEs was similar with both treatments (low SOE). More participants reported one or more AEs with combination treatment than monotherapy (moderate SOE).

3.7 Tolterodine plus ABs versus ABs alone

Four trials randomized men with LUTS and OAB symptoms ( n = 1338) to a combination of tolterodine 4 mg plus AB versus AB monotherapy with tamsulosin, doxazosin, or alfuzosin ( Table 5 ) 35 36 37 38 . Overall RoB for three trials was low 35 36 37 and one trial had a high RoB [38] .

Responder analysis was provided in only the high RoB trial [38] with a response defined as a 3-point improvement in I-PSS score from baseline ( Table 3 ). The proportion of responders was greater in the combination group than the AB monotherapy group (77% vs 29%), but the SOE for the comparison was judged as insufficient for this outcome due to a high RoB and unknown consistency. Mean changes in I-PSS scores were similar between treatment groups (WMD: –0.19, 95% CI: –1.08 to 0.68; moderate SOE). The mean change in I-PSS QoL was also not different between treatment groups (WMD: –0.34, 95% CI: –1.14 to 0.46; low SOE) 35 36 37 .

There were six incidences of AUR in the combination group versus two in the monotherapy group, which was judged as insufficient evidence 35 36 37 . Withdrawal for any reason (low SOE), withdrawal due to adverse effects (low SOE), and rates of ≥1 AEs (insufficient SOE) were not different between groups [35] .

3.8 Trospium plus ABs versus ABs alone

One 12-wk trial ( n = 58) compared trospium 45 mg daily doses with AB to AB monotherapy in men with LUTS and OAB symptoms attributed to BPH [39] . Individuals with a baseline postvoid residual of >100 ml were excluded. RoB was moderate.

Evidence was insufficient to assess efficacy for any outcome. One or more AE were reported in nine (35%) trospium participants versus five (23%) placebo patients.

3.9 Mirabegron plus AB versus AB alone

We found one 8-wk trial [40] ( n = 94), which compared 50 mg of mirabegron combined with 0.2 mg tamsulosin versus tamsulosin monotherapy in males with LUTS and OAB symptoms attributed to BPH. Patients with postvoid residual urine volume >100 ml were excluded. The study was judged to be at high RoB. The SOE was judged insufficient for all predetermined patient-important outcomes. We found no comparative trials combining mirabegron with other ABs for this indication.

3.10 Tadalafil versus tamsulosin

Four 3-mo trials compared tadalafil 2.5 mg, 5 mg, or 10 mg daily with tamsulosin 0.2 mg or 0.4 mg daily ( Table 1 ) 41 42 43 44 . Most participants had a history of erectile dysfunction (ED) [41 43 44] . The most frequently investigated dose level of tadalafil was 5 mg; one trial included a 2.5-mg dose level [42] and one trial evaluated 10 mg [41] . Overall RoB ranged from low to high for the four trials.

Tadalafil 5 mg and tamsulosin were similar in improving mean I-PSS scores (WMD: –0.07, 95% CI: –2.12 to 2.23; moderate SOE) and I-PSS QoL (WMD: –0.01, 95% CI: –0.75 to 0.73; low SOE). Evidence was insufficient for the outcomes of study withdrawal for any reason and the proportion of participants reporting at least one AE, but withdrawal due to AEs was higher with tadalafil (3% vs 1%; moderate SOE; Table 6 ).

3.11 Tadalafil versus alfuzosin

Two 3-mo trials ( n = 93) compared tadalafil with alfuzosin 10 mg daily ( Table 1 ) [45 46] . Kumar et al [45] compared tadalafil 10 mg daily with alfuzosin 10 mg daily. Liguori et al [46] compared tadalafil 20 mg taken on alternate days with alfuzosin 10 mg daily. All participants had a history of ED. Both trials were open label with a high overall RoB.

Alfuzosin 10 mg improved mean I-PSS scores more than tadalafil 10 mg or 20 mg (low SOE; Table 7 ). Mean reductions in I-PSS scores were 4.1 and 7.2 points with tadalafil and alfuzosin, respectively, favoring alfuzosin. I-PSS QoL also improved more with alfuzosin than tadalafil (low SOE). Study withdrawal for any reason and withdrawal due to an AE were no different with tadalafil or alfuzosin (insufficient SOE).

3.12 Sildenafil versus ABs

Three trials ( n = 181) compared sildenafil versus an AB 47 48 49 . One compared sildenafil 25 mg daily with alfuzosin 10 mg daily over 12 wk [48] and one compared sildenafil 25 mg taken 4 d/wk with tamsulosin 0.4 mg daily over 8 wk [49] . Abolyosr et al [47] compared sildenafil 50 mg to a low dose of doxazosin 2 mg over 16 wk; frequency of administration was not reported. All participants had a history of ED. All trials were open label and the overall RoB was high.

Mean change in I-PSS scores was –2.2 with sildenafil and –3.2 with alfuzosin or doxazosin (insufficient SOE) [47 48] . Mean change in I-PSS scores with sildenafil 25 mg was 4.0 points versus 5.4 points for tamsulosin 0.4 mg (insufficient SOE) [49] .

Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.13 Tadalafil AB combination versus AB monotherapy

Four trials ( n = 224) compared tadalafil combined with an AB versus AB monotherapy [41 45 46 50] . Two 3-mo trials compared tadalafil 10 mg daily [45] or 20 mg on alternate days [46] combined with alfuzosin 10 mg to alfuzosin 10 mg monotherapy. Two trials evaluated tadalafil combined with tamsulosin 0.4 mg versus tamsulosin 0.4 mg monotherapy: a 1-mo trial evaluated tadalafil 5 mg daily [50] and a 4-mo trial evaluated tadalafil 10 mg daily [41] . Nearly all participants had a history of ED [41 45 46] . All trials were open label except Regadas et al [50] and the overall RoB therefore ranged from moderate to high.

Results with combination therapy were similar to AB monotherapy (tadalafil 5–20 mg combined with AB was no different than AB monotherapy in improving mean I-PSS scores from baseline [WMD: –2.0, 95% CI: –4.03 to –0.00; insufficient SOE]). Mean reductions in I-PSS scores were similar; 10.4 and 8.6 with combination and monotherapy. Improvement in mean I-PSS QoL scores was also not significantly different between combination treatment and monotherapy; however, only open label (high RoB) trials reported this outcome (low SOE) [41 45 46] . Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.14 Sildenafil AB combination versus AB monotherapy

Four trials ( n = 281) compared sildenafil combined with an AB versus AB monotherapy 47 48 49 51] . Two 3-mo trials evaluated sildenafil combined with alfuzosin 10 mg versus alfuzosin monotherapy; one used daily sildenafil 25 mg [48] and the other used sildenafil 50 mg (dosing frequency not reported) [51] . One 4-mo trial evaluated sildenafil 50 mg combined with doxazosin 2 mg versus doxazosin alone, but dosing frequency was not reported [47] . An 8-wk trial evaluated sildenafil 25 mg taken 4 d/wk combined with tamsulosin 0.4 mg daily compared with tamsulosin monotherapy [49] . Three trials enrolled men with a history of ED 47 48 49 . All trials were open label or otherwise inadequately blinded and enrolled patients after they failed to respond to AB monotherapy. Overall RoB was mostly high.

Mean reductions in I-PSS scores were 5.4 with combination and 3.9 with monotherapy, with both treatments exceeding MDD. However, the SOE to assess the comparative effectiveness was insufficient due to the study limitations and imprecision in measurement [47 48 51] . In the 8-wk trial without data sufficient for pooling, improvement in mean I-PSS scores was similar with combination or monotherapy (–6.4 vs –5.4) [49] . Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.15 Vardenafil AB combination versus AB monotherapy

One double-blinded trial ( n = 60) compared vardenafil 10 mg daily combined with tamsulosin 0.4 mg to tamsulosin monotherapy over 12 wk [52] . The overall RoB was moderate. Mean reductions in I-PSS scores were 5.8 with combination treatment and 3.7 with monotherapy, both achieving MDD (insufficient SOE).

One withdrawal was reported with tamsulosin. No participant withdrew due to AEs. Persistent AEs were reported in three participants with combination therapy (headache with flushing, headache with stomach pain, and stomach pain) and two with tamsulosin (headache and flushing). No serious AEs were reported.

3.16 Tadalafil plus 5-ARI combination versus 5-ARI monotherapy and tadalafil versus 5-ARI/AB combination

The evidence from a single trial of tadalafil 5-ARI combination versus 5-ARI monotherapy [53] as well as that from a single trial comparing tadalafil versus 5-ARI/AB combination [54] was insufficient for both comparative effectiveness and safety.

3.17 Longer-term harms from observational studies

We identified two observational studies reporting longer-term AEs related to silodosin treatment [55 56] . In a 40-wk open label extension (cumulative treatment duration of 52 wk) of a previous RCT, 435 patients completed the extension in which a total of 431 experienced 924 AEs [55] . Twenty-nine patients (4.4%) experienced serious AEs including two deaths; none of the serious AEs were considered drug related, although no criteria were reported. The second study reviewed FDA data for AEs associated with ABs and found the evidence of silodosin insuffient to compare with other ABs [56] .

We identified one study examining long-term AEs associated with solifenacin/AB combination therapy, reporting an open extension from a previous RCT for a subset of patients with inclusion criteria that included a postvoid residual ≤150 ml [57] . Forty-seven percent of participants who continued solifenacin/AB combination treatment reported treatment-emergent AEs, most commonly dry mouth, constipation, and dyspepsia. In addition, 86 serious AEs occurred in 64 patients and included three deaths, six cases of AUR (0.7%), and three cases of intervertebral disc protrusion.

We found two longer-term observational studies on tadalafil [58 59] . In a 42-wk open label extension of a previous trial, 59% of 394 participants reported at least one AE and 9% withdrew due to an AE. Serious AEs were reported in 3% (11 participants) [58] . In another open extension study in a subset of 229 of 886 original participants, nearly 5% experienced serious AEs [59] .

2.1 Protocol

We developed an a priori written protocol, together with a technical report that incorporated input from key stakeholders, a multidisciplinary Technical Expert Panel, and public comment (available at the Agency for Healthcare Research and Quality [AHRQ] website http://effectivehealthcare.ahrq.gov/index.cfm/search-for-guides-reviews-and-reports/?productid=2067&pageaction=displayproduct ).

2.2 Eligibility criteria

Based on our Population, Interventions, Comparisons, Outcomes, Timing, and Setting criteria (Supplementary data) we included randomized controlled trials (RCTs) that tested comparative effectiveness of treatments involving newer drugs in men aged ≥45 yr with LUTS attributed to BPH. We defined these newer drugs as those that were Food and Drug Administration (FDA) approved for BPH since 2008 or which, though not FDA approved for BPH, have been studied for the treatment of BPH since 2008 and were selected through a formal process of stakeholder involvement (Supplementary data). Comparators included medications FDA approved for BPH before 2008. Included RCTs were at least 1 mo in duration with no minimum sample size. We additionally searched for large ( n ≥ 100 patients), longer-term (≥ 1 yr duration) observational studies to assess long-term or rare treatment associated harms only. We limited inclusion to English language articles.

The primary predefined outcomes of interest were changes reflecting clinically important differences (Supplementary data) in validated measures to assess LUTS (International Prostate Symptom Score [I-PSS]: score ranges 0–35 with higher scores indicating more severe symptoms; or American Urological Association Symptom Index scores), prostate-related bother or quality of life (QoL; I-PSS QoL question; BPH/LUTS impact scale), as well as rates of disease progression and/or treatment failure (prevention/delay of need for surgical intervention; acute urinary retention [AUR]). We also assessed common and serious medication adverse effects (AEs).

2.3 Information sources and literature search

We searched Ovid Medline, Ovid Embase, and the Cochrane Central Register of Controlled Trials with filters for study design (Supplementary data), to identify relevant RCTs published through June 20, 2016. We also searched for relevant systematic reviews and other key references. Lastly, we searched the Clinical Trials ( www.clinicaltrials.gov ) and the FDA ( www.fda.gov/Drugs ) websites to identify additional completed and ongoing studies.

2.4 Study selection process, data extraction, and risk of bias in studies

Two independent investigators screened titles and abstracts to identify studies meeting the eligibility criteria. Data were extracted by one investigator and reviewed and verified for accuracy by a second investigator. Risk of bias (RoB) of eligible studies was assessed using AHRQ guidance by one investigator and reviewed by a second [7] .

2.5 Synthesis of results

We assessed clinical and methodological heterogeneity and variation in effect size to determine the appropriateness of pooling data [8] . When three or more trials reported similar comparisons and outcomes, data were pooled using a Hartung, Knapp, Sidik, and Jonkman method [9] random effects model for proportion of I-PSS responders or mean changes in I-PSS scores in Stata (StataCorp., College Station, TX, USA). We pooled other outcomes in RevMan (RevMan, Spartanburg, SC, USA) [10] and converted DerSimonian-Laird random effects confidence intervals to Hartung, Knapp, Sidik, and Jonkman confidence intervals using an excel spreadsheet provided in Inthout et al [9] . We assessed between study variance with Tau ² and measured the magnitude of heterogeneity with the I ² statistic. If substantial heterogeneity was present (ie, I ² ≥ 70%), we stratified the results to assess treatment effects based on patient or study characteristics and/or explored sensitivity analyses [8 11] . We pooled across different ABs unless there were at least three trials for a given agent. We interpreted efficacy and comparative effectiveness using established thresholds indicating clinical significance (Supplementary data).

For the body of evidence from RCTs, we rated our confidence in the estimates of effect for the primary outcomes as overall strength of evidence (SOE) as high, moderate, low, or insufficient (Supplementary data) [12] . For observational studies, we did not formally assess SOE, but provided descriptive information in narrative form.

3.1 Search results

Our literature search identified 1139 references, of which 124 were selected for full-text review ( Fig. 1 ). This process mapped to 43 unique RCTs. In addition, we identified five longer duration (≥ 1 yr) observational studies reporting AEs.

 

Fig. 1
Figure 1: Literature flow diagram. The results are presented separately for each of four drug classes (new alpha-blockers, anticholinergics, beta-3 agonists, and phosphodiesterase type-5 inhibitor), and specific drugs are listed within each class. The outcomes addressed by the three key questions are discussed within each drug-specific section.

BPH = benign prostatic hyperplasia; RCT = randomized controlled trial.

3.2 Silodosin versus tamsulosin

Ten trials randomized men with LUTS attributed to BPH ( n = 1799) to silodosin 8 mg daily versus tamsulosin 0.2–0.4 mg daily 13 14 15 16 17 18 19 20 21 22 . Table 1 provides baseline characteristics. Overall, the RoB was low in two trials [13 18] , moderate in six trials 14 15 16 20 21 22 , and high in two trials [17 19] .

Three trials conducted responder analyses (defined as > 25% reduction in I-PSS score; Table 2 ) [16 18 20] . Response to treatment was not significantly different between silodosin and tamsulosin (risk ratio: 1.07, 95% confidence interval [CI]: 0.91–1.26; moderate SOE). Silodosin and tamsulosin (all dose levels) were not significantly different in improving mean I-PSS scores (weighted mean difference [WMD]: –0.52, 95% CI: –1.58 to 0.54; moderate SOE). Results were similar in analyses restricted to trials using tamsulosin 0.4 mg.

Overall withdrawals (for any reason) and rates of participants with ≥1 AEs were not significantly different between treatments (low and moderate SOE, respectively). Withdrawals due to AEs were more frequent for silodosin versus tamsulosin (risk ratio: 1.96, 95% CI: 1.04–3.71; moderate SOE). The most common AE was abnormal ejaculation, reported in 16% of patients on silodosin versus 2% of patients on tamsulosin.

3.3 Darifenacin plus ABs versus ABs alone

Two 146 12-wk trials ( n = 161) compared darifenacin/AB combination 147 therapy with AB monotherapy in men with LUTS and overactive bladder (OAB) symptoms attributed to BPH [23 24] . Participants with a baseline postvoid residual of >150 ml were excluded. RoB was low in one trial [24] and moderate in the other [23] . SOE was judged insufficient for all outcomes.

3.4 Fesoterodine plus ABs versus ABs alone

Two trials ( n = 990) compared fesoterodine/AB combination therapy with AB monotherapy ( Table 1 ) in men with LUTS and OAB symptoms [25 26] . Overall RoB was moderate for one trial [25] and high for the other [26] . Improvement in mean I-PSS scores was similar with fesoterodine/AB combination and AB monotherapy (low SOE; Table 3 ). The mean difference in the large moderate RoB trial was 0.0 (95% CI: –0.83 to 0.83) and –1.70 (95% CI: –5.85 to 2.46) for the small high RoB trial. Overall withdrawals, withdrawals due to AEs, and the number of participants with ≥1 AE were more frequent in the fesoterodine arm; SOE was judged as low for all three outcomes.

3.5 Oxybutynin plus ABs versus ABs alone

One 12-wk trial ( n = 420) trial compared oxybutynin 10 mg tablets plus tamsulosin 0.4 mg with tamsulosin 0.4 mg alone [27] . Individuals with a baseline postvoid residual of >200 ml were excluded. RoB was moderate. SOE was judged insufficient for all outcomes.

3.6 Solifenacin plus ABs versus ABs alone

Seven 12-wk trials 28 29 30 31 32 33 34 randomized men with LUTS and OAB symptoms ( n = 3147) to solifenacin plus tamsulosin versus tamsulosin monotherapy ( Table 1 ). Five trials examined solifenacin 5 mg 28 29 30 31 34] and two examined solifenacin 6 mg [32 33] . The overall RoB was moderate.

Combination therapy was similar to AB monotherapy in improving LUTS (moderate SOE; Table 4 ). Improvement in mean I-PSS score from baseline was similar with solifenacin 5 mg or 6 mg plus tamsulosin 0.2 mg or 0.4 mg versus tamsulosin alone (WMD: –0.29, 95% CI: –0.88 to 0.30; moderate SOE). Combination therapy lowered I-PSS QoL score more than tamsulosin, but the difference between groups was not clinically significant based on predetermined thresholds (moderate SOE). Evidence from four trials using solifenacin 3–9 mg [29 32 33 34 reported no statistical difference between treatment groups in rates of AUR, but evidence was judged insufficient to draw conclusions due to imprecision.

Withdrawal for any reason or due to AEs was similar with both treatments (low SOE). More participants reported one or more AEs with combination treatment than monotherapy (moderate SOE).

3.7 Tolterodine plus ABs versus ABs alone

Four trials randomized men with LUTS and OAB symptoms ( n = 1338) to a combination of tolterodine 4 mg plus AB versus AB monotherapy with tamsulosin, doxazosin, or alfuzosin ( Table 5 ) 35 36 37 38 . Overall RoB for three trials was low 35 36 37 and one trial had a high RoB [38] .

Responder analysis was provided in only the high RoB trial [38] with a response defined as a 3-point improvement in I-PSS score from baseline ( Table 3 ). The proportion of responders was greater in the combination group than the AB monotherapy group (77% vs 29%), but the SOE for the comparison was judged as insufficient for this outcome due to a high RoB and unknown consistency. Mean changes in I-PSS scores were similar between treatment groups (WMD: –0.19, 95% CI: –1.08 to 0.68; moderate SOE). The mean change in I-PSS QoL was also not different between treatment groups (WMD: –0.34, 95% CI: –1.14 to 0.46; low SOE) 35 36 37 .

There were six incidences of AUR in the combination group versus two in the monotherapy group, which was judged as insufficient evidence 35 36 37 . Withdrawal for any reason (low SOE), withdrawal due to adverse effects (low SOE), and rates of ≥1 AEs (insufficient SOE) were not different between groups [35] .

3.8 Trospium plus ABs versus ABs alone

One 12-wk trial ( n = 58) compared trospium 45 mg daily doses with AB to AB monotherapy in men with LUTS and OAB symptoms attributed to BPH [39] . Individuals with a baseline postvoid residual of >100 ml were excluded. RoB was moderate.

Evidence was insufficient to assess efficacy for any outcome. One or more AE were reported in nine (35%) trospium participants versus five (23%) placebo patients.

3.9 Mirabegron plus AB versus AB alone

We found one 8-wk trial [40] ( n = 94), which compared 50 mg of mirabegron combined with 0.2 mg tamsulosin versus tamsulosin monotherapy in males with LUTS and OAB symptoms attributed to BPH. Patients with postvoid residual urine volume >100 ml were excluded. The study was judged to be at high RoB. The SOE was judged insufficient for all predetermined patient-important outcomes. We found no comparative trials combining mirabegron with other ABs for this indication.

3.10 Tadalafil versus tamsulosin

Four 3-mo trials compared tadalafil 2.5 mg, 5 mg, or 10 mg daily with tamsulosin 0.2 mg or 0.4 mg daily ( Table 1 ) 41 42 43 44 . Most participants had a history of erectile dysfunction (ED) [41 43 44] . The most frequently investigated dose level of tadalafil was 5 mg; one trial included a 2.5-mg dose level [42] and one trial evaluated 10 mg [41] . Overall RoB ranged from low to high for the four trials.

Tadalafil 5 mg and tamsulosin were similar in improving mean I-PSS scores (WMD: –0.07, 95% CI: –2.12 to 2.23; moderate SOE) and I-PSS QoL (WMD: –0.01, 95% CI: –0.75 to 0.73; low SOE). Evidence was insufficient for the outcomes of study withdrawal for any reason and the proportion of participants reporting at least one AE, but withdrawal due to AEs was higher with tadalafil (3% vs 1%; moderate SOE; Table 6 ).

3.11 Tadalafil versus alfuzosin

Two 3-mo trials ( n = 93) compared tadalafil with alfuzosin 10 mg daily ( Table 1 ) [45 46] . Kumar et al [45] compared tadalafil 10 mg daily with alfuzosin 10 mg daily. Liguori et al [46] compared tadalafil 20 mg taken on alternate days with alfuzosin 10 mg daily. All participants had a history of ED. Both trials were open label with a high overall RoB.

Alfuzosin 10 mg improved mean I-PSS scores more than tadalafil 10 mg or 20 mg (low SOE; Table 7 ). Mean reductions in I-PSS scores were 4.1 and 7.2 points with tadalafil and alfuzosin, respectively, favoring alfuzosin. I-PSS QoL also improved more with alfuzosin than tadalafil (low SOE). Study withdrawal for any reason and withdrawal due to an AE were no different with tadalafil or alfuzosin (insufficient SOE).

3.12 Sildenafil versus ABs

Three trials ( n = 181) compared sildenafil versus an AB 47 48 49 . One compared sildenafil 25 mg daily with alfuzosin 10 mg daily over 12 wk [48] and one compared sildenafil 25 mg taken 4 d/wk with tamsulosin 0.4 mg daily over 8 wk [49] . Abolyosr et al [47] compared sildenafil 50 mg to a low dose of doxazosin 2 mg over 16 wk; frequency of administration was not reported. All participants had a history of ED. All trials were open label and the overall RoB was high.

Mean change in I-PSS scores was –2.2 with sildenafil and –3.2 with alfuzosin or doxazosin (insufficient SOE) [47 48] . Mean change in I-PSS scores with sildenafil 25 mg was 4.0 points versus 5.4 points for tamsulosin 0.4 mg (insufficient SOE) [49] .

Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.13 Tadalafil AB combination versus AB monotherapy

Four trials ( n = 224) compared tadalafil combined with an AB versus AB monotherapy [41 45 46 50] . Two 3-mo trials compared tadalafil 10 mg daily [45] or 20 mg on alternate days [46] combined with alfuzosin 10 mg to alfuzosin 10 mg monotherapy. Two trials evaluated tadalafil combined with tamsulosin 0.4 mg versus tamsulosin 0.4 mg monotherapy: a 1-mo trial evaluated tadalafil 5 mg daily [50] and a 4-mo trial evaluated tadalafil 10 mg daily [41] . Nearly all participants had a history of ED [41 45 46] . All trials were open label except Regadas et al [50] and the overall RoB therefore ranged from moderate to high.

Results with combination therapy were similar to AB monotherapy (tadalafil 5–20 mg combined with AB was no different than AB monotherapy in improving mean I-PSS scores from baseline [WMD: –2.0, 95% CI: –4.03 to –0.00; insufficient SOE]). Mean reductions in I-PSS scores were similar; 10.4 and 8.6 with combination and monotherapy. Improvement in mean I-PSS QoL scores was also not significantly different between combination treatment and monotherapy; however, only open label (high RoB) trials reported this outcome (low SOE) [41 45 46] . Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.14 Sildenafil AB combination versus AB monotherapy

Four trials ( n = 281) compared sildenafil combined with an AB versus AB monotherapy 47 48 49 51] . Two 3-mo trials evaluated sildenafil combined with alfuzosin 10 mg versus alfuzosin monotherapy; one used daily sildenafil 25 mg [48] and the other used sildenafil 50 mg (dosing frequency not reported) [51] . One 4-mo trial evaluated sildenafil 50 mg combined with doxazosin 2 mg versus doxazosin alone, but dosing frequency was not reported [47] . An 8-wk trial evaluated sildenafil 25 mg taken 4 d/wk combined with tamsulosin 0.4 mg daily compared with tamsulosin monotherapy [49] . Three trials enrolled men with a history of ED 47 48 49 . All trials were open label or otherwise inadequately blinded and enrolled patients after they failed to respond to AB monotherapy. Overall RoB was mostly high.

Mean reductions in I-PSS scores were 5.4 with combination and 3.9 with monotherapy, with both treatments exceeding MDD. However, the SOE to assess the comparative effectiveness was insufficient due to the study limitations and imprecision in measurement [47 48 51] . In the 8-wk trial without data sufficient for pooling, improvement in mean I-PSS scores was similar with combination or monotherapy (–6.4 vs –5.4) [49] . Evidence was insufficient for overall withdrawals and withdrawals due to AEs.

3.15 Vardenafil AB combination versus AB monotherapy

One double-blinded trial ( n = 60) compared vardenafil 10 mg daily combined with tamsulosin 0.4 mg to tamsulosin monotherapy over 12 wk [52] . The overall RoB was moderate. Mean reductions in I-PSS scores were 5.8 with combination treatment and 3.7 with monotherapy, both achieving MDD (insufficient SOE).

One withdrawal was reported with tamsulosin. No participant withdrew due to AEs. Persistent AEs were reported in three participants with combination therapy (headache with flushing, headache with stomach pain, and stomach pain) and two with tamsulosin (headache and flushing). No serious AEs were reported.

3.16 Tadalafil plus 5-ARI combination versus 5-ARI monotherapy and tadalafil versus 5-ARI/AB combination

The evidence from a single trial of tadalafil 5-ARI combination versus 5-ARI monotherapy [53] as well as that from a single trial comparing tadalafil versus 5-ARI/AB combination [54] was insufficient for both comparative effectiveness and safety.

3.17 Longer-term harms from observational studies

We identified two observational studies reporting longer-term AEs related to silodosin treatment [55 56] . In a 40-wk open label extension (cumulative treatment duration of 52 wk) of a previous RCT, 435 patients completed the extension in which a total of 431 experienced 924 AEs [55] . Twenty-nine patients (4.4%) experienced serious AEs including two deaths; none of the serious AEs were considered drug related, although no criteria were reported. The second study reviewed FDA data for AEs associated with ABs and found the evidence of silodosin insuffient to compare with other ABs [56] .

We identified one study examining long-term AEs associated with solifenacin/AB combination therapy, reporting an open extension from a previous RCT for a subset of patients with inclusion criteria that included a postvoid residual ≤150 ml [57] . Forty-seven percent of participants who continued solifenacin/AB combination treatment reported treatment-emergent AEs, most commonly dry mouth, constipation, and dyspepsia. In addition, 86 serious AEs occurred in 64 patients and included three deaths, six cases of AUR (0.7%), and three cases of intervertebral disc protrusion.

We found two longer-term observational studies on tadalafil [58 59] . In a 42-wk open label extension of a previous trial, 59% of 394 participants reported at least one AE and 9% withdrew due to an AE. Serious AEs were reported in 3% (11 participants) [58] . In another open extension study in a subset of 229 of 886 original participants, nearly 5% experienced serious AEs [59] .

2.1 Protocol

We developed an a priori written protocol, together with a technical report that incorporated input from key stakeholders, a multidisciplinary Technical Expert Panel, and public comment (available at the Agency for Healthcare Research and Quality [AHRQ] website http://effectivehealthcare.ahrq.gov/index.cfm/search-for-guides-reviews-and-reports/?productid=2067&pageaction=displayproduct ).

2.2 Eligibility criteria

Based on our Population, Interventions, Comparisons, Outcomes, Timing, and Setting criteria (Supplementary data) we included randomized controlled trials (RCTs) that tested comparative effectiveness of treatments involving newer drugs in men aged ≥45 yr with LUTS attributed to BPH. We defined these newer drugs as those that were Food and Drug Administration (FDA) approved for BPH since 2008 or which, though not FDA approved for BPH, have been studied for the treatment of BPH since 2008 and were selected through a formal process of stakeholder involvement (Supplementary data). Comparators included medications FDA approved for BPH before 2008. Included RCTs were at least 1 mo in duration with no minimum sample size. We additionally searched for large ( n ≥ 100 patients), longer-term (≥ 1 yr duration) observational studies to assess long-term or rare treatment associated harms only. We limited inclusion to English language articles.

The primary predefined outcomes of interest were changes reflecting clinically important differences (Supplementary data) in validated measures to assess LUTS (International Prostate Symptom Score [I-PSS]: score ranges 0–35 with higher scores indicating more severe symptoms; or American Urological Association Symptom Index scores), prostate-related bother or quality of life (QoL; I-PSS QoL question; BPH/LUTS impact scale), as well as rates of disease progression and/or treatment failure (prevention/delay of need for surgical intervention; acute urinary retention [AUR]). We also assessed common and serious medication adverse effects (AEs).

2.3 Information sources and literature search

We searched Ovid Medline, Ovid Embase, and the Cochrane Central Register of Controlled Trials with filters for study design (Supplementary data), to identify relevant RCTs published through June 20, 2016. We also searched for relevant systematic reviews and other key references. Lastly, we searched the Clinical Trials ( www.clinicaltrials.gov ) and the FDA ( www.fda.gov/Drugs ) websites to identify additional completed and ongoing studies.

2.4 Study selection process, data extraction, and risk of bias in studies

Two independent investigators screened titles and abstracts to identify studies meeting the eligibility criteria. Data were extracted by one investigator and reviewed and verified for accuracy by a second investigator. Risk of bias (RoB) of eligible studies was assessed using AHRQ guidance by one investigator and reviewed by a second [7] .

2.5 Synthesis of results

We assessed clinical and methodological heterogeneity and variation in effect size to determine the appropriateness of pooling data [8] . When three or more trials reported similar comparisons and outcomes, data were pooled using a Hartung, Knapp, Sidik, and Jonkman method [9] random effects model for proportion of I-PSS responders or mean changes in I-PSS scores in Stata (StataCorp., College Station, TX, USA). We pooled other outcomes in RevMan (RevMan, Spartanburg, SC, USA) [10] and converted DerSimonian-Laird random effects confidence intervals to Hartung, Knapp, Sidik, and Jonkman confidence intervals using an excel spreadsheet provided in Inthout et al [9] . We assessed between study variance with Tau ² and measured the magnitude of heterogeneity with the I ² statistic. If substantial heterogeneity was present (ie, I ² ≥ 70%), we stratified the results to assess treatment effects based on patient or study characteristics and/or explored sensitivity analyses [8 11] . We pooled across different ABs unless there were at least three trials for a given agent. We interpreted efficacy and comparative effectiveness using established thresholds indicating clinical significance (Supplementary data).

For the body of evidence from RCTs, we rated our confidence in the estimates of effect for the primary outcomes as overall strength of evidence (SOE) as high, moderate, low, or insufficient (Supplementary data) [12] . For observational studies, we did not formally assess SOE, but provided descriptive information in narrative form.

3.1 Search results

Our literature search identified 1139 references, of which 124 were selected for full-text review ( Fig. 1 ). This process mapped to 43 unique RCTs. In addition, we identified five longer duration (≥ 1 yr) observational studies reporting AEs.

 

Fig. 1
Figure 1: Literature flow diagram. The results are presented separately for each of four drug classes (new alpha-blockers, anticholinergics, beta-3 agonists, and phosphodiesterase type-5 inhibitor), and specific drugs are listed within each class. The outcomes addressed by the three key questions are discussed within each drug-specific section.

BPH = benign prostatic hyperplasia; RCT = randomized controlled trial.

3.2 Silodosin versus tamsulosin

Ten trials randomized men with LUTS attributed to BPH ( n = 1799) to silodosin 8 mg daily versus tamsulosin 0.2–0.4 mg daily 13 14 15 16 17 18 19 20 21 22 . Table 1 provides baseline characteristics. Overall, the RoB was low in two trials [13 18] , moderate in six trials 14 15 16 20 21 22 , and high in two trials [17 19] .

Three trials conducted responder analyses (defined as > 25% reduction in I-PSS score; Table 2 ) [16 18 20] . Response to treatment was not significantly different between silodosin and tamsulosin (risk ratio: 1.07, 95% confidence interval [CI]: 0.91–1.26; moderate SOE). Silodosin and tamsulosin (all dose levels) were not significantly different in improving mean I-PSS scores (weighted mean difference [WMD]: –0.52, 95% CI: –1.58 to 0.54; moderate SOE). Results were similar in analyses restricted to trials using tamsulosin 0.4 mg.