Twenty one RCTs (n=4199). Seven were Phase III trials and 14 were Phase II trials.
Penile rigidity (Phase II trials only, n=7): In all studies an increased duration of rigidity >60% was seen with increasing doses of sildenafil.
Primary end points:
Frequency of penetration: 9/9 RCTs showed statistically significant improvements with all doses of sildenafil compared to placebo. Increasing improvement was apparent with increasing dose over the range 25-100mg.
Maintenance of erection after penetration: 9/9 RCTs showed a statistically significant improvement with all doses of sildenafil compared to placebo. Increasing improvement was apparent with increasing dose over the range 25-100mg.
Secondary end points:
General efficacy question: 21/21 RCTs reported an improvement in erections with sildenafil treatment (65-82% improved) compared to placebo (18-26% improved). A dose-response relationship was seen over the dose range 25-100mg.
Successful intercourse: Proportion of attempts at intercourse that were successful with sildenafil treatment increased compared to placebo treatment (6/9 RCTs list a positive outcome). One RCT found no increase in rates of attempted intercourse with sildenafil, despite increased success rates.
Erections hard enough for intercourse: In 6/6 RCTs a significant improvement in the number of grade 3 and 4 erections was seen with sildenafil treatment. A dose-response relationship was apparent.
Partner's questionnaire: Generally increasing partner satisfaction was seen with increasing sildenafil dose (10-100mg).
Quality of life: Statistically significant but small quality of life treatment effects were seen in 4 RCTs.
Adverse effects: Adverse events reported in >2% (range 2% - 10%) of patient treated (2RCTs) include headache, flushing, dyspepsia, nasal congestion, urinary tract infection, abnormal vision, diarrhoea, dizziness, rash, respiratory tract infection. No cases of priapism were reported in any of the sildenafil studies.
Cardiovascular: 18 RCTs give an overall incidence of cardiovascular events (other than flushing) of 3.0% with sildenafil and 3.5% with placebo. A comparable incidence rate was seen with sildenafil and placebo for serious cardiovascular events. No clinically significant changes in blood pressure, heart rate or ECG were seen with sildenafil treatment. Sildenafil had no direct effect on platelet aggregation.
Death: 8/4500 patients treated with sildenafil in clinical trial died whilst taking the study drug or within 30 days of treatment. Six deaths were related to cardiovascular causes; in all cases it was plausible that the event was not related to the study drug.
Diabetics: In general, smaller improvements were seen in diabetics than in those without diabetes.
Spinal cord injury: sildenafil appears to show comparable efficacy in patients with erectile dysfunction solely attributable to spinal cord injury (but with intact spinal cord-mediated reflexes) to patients with erectile dysfunction of broad-spectrum aetiology.
Radical prostatectomy: 4% of participants had erectile dysfunction as a result of radical prostatectomy. A subgroup analysis of the patients appeared ot show lower efficacy with sildenafil, with 40-50% achieving improved erections.
Age: Age did not affect response to sildenafil.