Twenty three RCTs (n=2,315) were included in the review. Lowest risk of bias was for selective outcome reporting (22 RCTs) and blinding (20 RCTs). Greatest risk of bias was for allocation concealment, which was adequate for only five studies.
Alpha-blockers were significantly more effective in reducing total symptom scores compared to placebo (SMD -1.7, 95% CI -2.8 to -0.6; five RCTs). There was evidence of statistical heterogeneity (I2=96.4%), the source of which was not identified using meta-regression. There was no statistically significant difference between antibiotics and placebo (three RCTs). Sensitivity analysis for alpha-blockers versus placebo did not significantly alter the results.
Alpha-blockers and phytotherapy were statistically significantly more effective than placebo in reducing pain scores (SMD -1.1, 95% CI -1.8 to -0.3, I2=94%; six RCTs for alpha-blockers and SMD -0.5, 95% CI -0.7 to -0.2; three RCTs for phytotherapy). There were no significant differences between antibiotics and placebo (three RCTs). Sensitivity analysis for alpha-blockers versus placebo did not significantly alter the results.
Alpha-blockers, antibiotics and phytotherapy were statistically significantly more effective than placebo in reducing voiding (SMD -1.4, 95% CI -2.3 to -0.5, I2=95.5%; five RCTs for alpha-blockers and SMD -3.2, 95% CI -6.1 to -0.3; three RCTs for antibiotics and SMD -0.4, 95% CI -0.7 to -0.1; three RCTs for phytotherapy). Sensitivity analysis for alpha-blockers versus placebo did not significantly alter the findings.
Alpha-blockers statistically significantly improved quality of life compared to placebo (SMD -1.0, 95% CI -1.8 to -0.2, I2=94.5%; five RCTs). There was no statistically significant difference in quality of life scores between antibiotics and placebo (three RCTs).
There was evidence of publication bias for alpha-blockers on all outcomes and adjusting for this resulted in non-significant treatment effects.
Alpha-blockers and anti-inflammatories were statistically significantly more likely to have a favourable treatment response than placebo (RR 1.6, 95% CI 1.1 to 2.3; six RCTs for alpha-blockers and RR 1.8, 95% CI 1.2 to 2.6, I2=24.4%; three RCTs for anti-inflammatories). There was evidence of statistical heterogeneity (I2=64.2%) for alpha-blockers, but this was no longer significant when adjusting for duration of treatment using meta-regression.
Network meta-analysis showed that alpha-blockers, antibiotics and alpha-blockers plus antibiotics were more effective than placebo for all outcomes. Finasteride was more effective than placebo for total symptom scores. Anti-inflammatories were more effective than placebo for response rates. Alpha-blockers plus antibiotics was the most effective treatment for total symptom scores, pain, voiding and quality of life scores; anti-inflammatories were the most effective for response rates.
Further results were reported in the review.