Peptide & Experimental
Viagra (Sildenafil)
Phosphodiesterase-5 inhibitor (PDE5 inhibitor), prescription-only medicine · Viagra, Revatio, sildenafil generics, PDE5 inhibitor
Sildenafil reliably improves erectile function and is among the best-supported drugs of all for this indication. The PDE5 inhibitor has been approved since 1998. The proposed protective effect against dementia, by contrast, is contradictory and open.
What sildenafil is
Sildenafil is a prescription-only phosphodiesterase-5 inhibitor, PDE5 inhibitor for short. In the European Union, sildenafil has been approved as Viagra since 1998 for erectile dysfunction and as Revatio since 2005 for pulmonary arterial hypertension.
What is rated here is the state of knowledge. For erectile dysfunction, the evidence is exceptionally dense: meta-analyses of dozens of randomized trials with several thousand participants, plus decades of post-marketing surveillance.
How it works
During sexual arousal, nerve endings release nitric oxide, which triggers the formation of the messenger cGMP. cGMP relaxes the smooth vascular muscle in the erectile tissue, and blood flows in. The enzyme PDE5 breaks down cGMP; sildenafil inhibits it and prolongs the signal. Without arousal, nothing happens.
Because PDE5 is found not only in the erectile tissue, further uses are being studied. A plausible mechanism, however, is not proof of efficacy for another indication.
What is well supported
- An exceptionally low NNT. In the meta-analysis by Moore 2002 of 10 randomized trials with 2,123 men on treatment and 1,131 on placebo, 49 versus 11 percent achieved at least 60 percent successful attempts. This gives a number needed to treat of 2.7 (2.3 to 3.3), and for global improvement even 1.7 (1.6 to 1.9). Fewer than three men therefore need to be treated for one to benefit beyond the placebo effect.
- The effect has been reproduced in large numbers. Fink 2002 analyzed 27 randomized trials with 6,659 men with erectile dysfunction. The share of successful attempts was 57 versus 21 percent on placebo, weighted mean difference 33.7 (29.2 to 38.2). At least one successful attempt was reported by 83 versus 45 percent, relative benefit increase 1.8 (1.7 to 1.9).
- Serious side effects were no more frequent than on placebo. The harms meta-analysis by Tsertsvadze 2009 of 49 randomized trials did find more side effects overall (RR 1.56, 1.38 to 1.76, dose-dependent), but serious side effects and study discontinuations were not more frequent than on placebo. Fink 2002 also found no significant association with serious cardiovascular events or death.
- Approved since 1998. The European Medicines Agency approved Viagra for erectile dysfunction on September 14, 1998. For the approved indication, there is thus more than a quarter of a century of post-marketing surveillance, from which a clearly defined warning profile has emerged.
What the studies show
Meta-analysis on efficacy
Fink 2002 (Arch Intern Med) analyzed 27 randomized trials with 6,659 men with erectile dysfunction. The share of successful attempts was 57 percent versus 21 percent on placebo, weighted mean difference 33.7 (95 percent confidence interval 29.2 to 38.2; n=2,283). At least one successful attempt was reported by 83 versus 45 percent, relative benefit increase 1.8 (1.7 to 1.9). Serious cardiovascular events or deaths were not significantly more frequent.
How many need to be treated per success
Moore 2002 (BMC Urol) pooled 10 randomized trials with 2,123 treated and 1,131 placebo participants. At least 60 percent successful attempts were achieved by 49 versus 11 percent, which gives a number needed to treat of 2.7 (2.3 to 3.3). For global improvement it was 1.7 (1.6 to 1.9). Treatment-related side effects occurred in 30 versus 11 percent, number needed to harm 5.4 (4.3 to 7.3).
The harms profile
Tsertsvadze 2009 (Urology) analyzed 49 randomized trials specifically for harms. Side effects overall were more frequent, relative risk 1.56 (1.38 to 1.76), with a dose-dependent increase. Serious side effects and study discontinuations, by contrast, were not more frequent than on placebo. Long-term data are lacking; Fink 2002 does not cover safety beyond six months.
The dementia debate: open, not settled
The findings contradict each other. Chua 2025 (Aging) found, across 5 observational studies with 885,380 people, a hazard ratio of 0.47 (0.27 to 0.82), that is, a markedly lower risk. Gronich 2025 (Neuroepidemiology), by contrast, found no effect in 133,336 patients with erectile dysfunction over 7.9 years, hazard ratio 0.95 (0.86 to 1.04). A Mendelian randomization even gave an odds ratio of 1.09 (1.07 to 1.11) for Alzheimer’s, that is, the opposite direction. This is an open question, not a finding.
Where the data stop
- Protection against dementia or Alzheimer’s. One large cohort analysis found a markedly lower risk, a second found no effect at all, and a Mendelian randomization pointed in the opposite direction. Contradictory observational data do not amount to proof of efficacy.
- Benefit in heart failure with preserved ejection fraction. Hoendermis 2015 (Eur Heart J) studied 52 HFpEF patients over 12 weeks. The primary endpoint, mean pulmonary artery pressure, fell by 2.4 versus 4.7 mmHg on placebo, p=0.14. The trial missed its goal.
- A general circulation or anti-aging benefit. The approvals cover erectile dysfunction and, as Revatio, pulmonary arterial hypertension. There are no approval data for marketing it as a general vascular drug.
- Long-term safety. The available meta-analyses do not extend beyond six months. Statements about use over years cannot be derived from them.
Status, approval and legal
Viagra received EU approval for erectile dysfunction on September 14, 1998, and Revatio in 2005 for pulmonary arterial hypertension. The drug is prescription-only.
It is contraindicated in concomitant use with nitrates and with riociguat, as well as in severe heart disease and with a history of non-arteritic anterior ischemic optic neuropathy (NAION), a circulatory disorder of the optic nerve.
Safety
The side-effect profile has been quantified: in the meta-analysis by Fink 2002, flushing occurred in 12 percent, headache in 11 percent, dyspepsia in 5 percent and visual disturbances in 3 percent. Overall, side effects are more frequent than on placebo, relative risk 1.56, with a dose-dependent increase.
The other side of this balance: serious side effects and treatment discontinuations were not more frequent than on placebo in the harms meta-analysis of 49 trials, and no significant association with serious cardiovascular events or death was found. This applies to periods of at most six months and does not replace the contraindications.
BK-Score Well supported
| Human evidence | 10 | |
|---|---|---|
| Mechanism | 9 | |
| Safety data | 9 | |
| Hype gap | 7 | |
| Track record of use | 10 |
Approved since 1998 on the basis of the approval trials around Goldstein et al. (NEJM 1998), with a significant improvement in erectile function versus placebo, later additionally for pulmonary arterial hypertension. PDE5 inhibition is among the best-characterized chains of effects in pharmacology. More than 25 years of post-marketing surveillance provide a complete risk profile, including the nitrate contraindication. Increasingly, a general circulation and anti-aging benefit is also advertised – there are no approval data for that.
The score rates the state of knowledge, not the substance. “Safety data 9” means well studied – not harmless. “Track record of use 9” means used long and widely – that is not proof of efficacy.
Subjective assessment by Biohacking Kompakt based on published scoring rules – not a scientific rating and not a medical recommendation. Rules and all ratings (German)
Frequently asked questions about Viagra (Sildenafil)
Does sildenafil really work?
For the approved indication, yes, and clearly so. Across 27 randomized trials with 6,659 men, the share of successful attempts was 57 percent versus 21 percent on placebo. The number needed to treat is 2.7, which means that, mathematically, about one in three of those treated benefits compared with placebo.
Does sildenafil protect against dementia?
That is open, and the data contradict each other. An analysis of five observational studies with 885,380 people found a markedly lower risk with a hazard ratio of 0.47. A second study with 133,336 patients found no effect, and a genetic analysis even pointed in the opposite direction.
What are the side effects of sildenafil?
In the 2002 meta-analysis, facial flushing occurred in 12 percent, headache in 11 percent, digestive complaints in 5 percent and visual disturbances in 3 percent. Side effects overall were more frequent than on placebo, relative risk 1.56. Serious side effects and discontinuations, by contrast, were not more frequent.
Is sildenafil dangerous for the heart?
The meta-analyses found no significant association with serious cardiovascular events or death. The contraindications remain unaffected by this: concomitant use with nitrates or riociguat is ruled out, as is use in severe heart disease. Long-term data beyond six months are lacking.
Does sildenafil help with heart failure or pulmonary hypertension?
For heart failure with preserved ejection fraction, the answer is no. In a trial with 52 patients over 12 weeks, mean pulmonary artery pressure fell by 2.4 versus 4.7 mmHg on placebo, p = 0.14. The primary endpoint was thus missed.
Is sildenafil approved in Germany?
Yes. Viagra received EU approval for erectile dysfunction on September 14, 1998, and Revatio in 2005 for pulmonary arterial hypertension. The drug is prescription-only. Contraindications include concomitant use of nitrates or riociguat, severe heart disease and a history of NAION, a circulatory disorder of the optic nerve.
The podcast episode (in German)
Episode 44
Viagra (Sildenafil): the blue pill fact-checked
The podcast by Paul Höser (Episode 44) · with Paul & Paula. The whole story of the world’s most famous pill: from the failed heart drug UK-92480 via the Welsh miners and the approval trials (Boolell 1996; Goldstein, NEJM 1998) to the Nobel Prize mechanism (NO–cGMP–PDE5) and the longevity turn: −69 % Alzheimer’s risk in the Cleveland Clinic analysis (Nature Aging 2021), −18–44 % in the UCL study (Brauer, Neurology 2024) – honestly put into context. Plus the erection as an early vascular warning system, altitude medicine, the nitrate rule and fun facts from cut flowers to hamster jet lag. Information only, no dosing or usage recommendation – prescription-only.
Related
- Same substance classTadalafil
- Related topicSemaglutide (Ozempic / Wegovy)
- Same sectionPT-141
- Same sectionTirzepatide (Mounjaro / Zepbound)
- Same sectionHGH (Growth Hormone / Somatropin)
- ComparisonTadalafil vs. Viagra (Sildenafil)
Sources
- Fink HA et al. 2002, Arch Intern Med - meta-analysis of 27 RCTs, efficacy (PMID 12076233)
- Moore RA et al. 2002, BMC Urol - 10 RCTs, NNT and NNH (PMID 12049673)
- Tsertsvadze A et al. 2009, Urology - harms meta-analysis of 49 RCTs (PMID 19592078)
- Chua WY et al. 2025, Aging - 5 observational studies, dementia risk (PMID 40096550)
- Gronich N et al. 2025, Neuroepidemiology - cohort, no effect on dementia (PMID 38952132)
- Mendelian randomization, PDE5 inhibition and Alzheimer’s (PMID 39951190)
- Hoendermis ES et al. 2015, Eur Heart J - HFpEF, primary endpoint missed (PMID 26188003)
- EMA - Viagra, European public assessment report
Open in the database – with search, filters and comparison (German app)
Information only, not medical advice and no usage or dosing recommendation. Prescription-only and unapproved substances belong in medical hands. Last updated: 2026-10-04.