It no longer works the way it used to, and the first suspicion falls on testosterone. That is understandable, because every advertisement puts erections and testosterone in the same sentence. In many cases it is the wrong suspicion, and it costs time.
Erectile function and testosterone are two different systems that overlap without being the same thing. Confusing them means either treating the wrong thing or missing something more important than both.
Two different machines
An erection is primarily a matter of blood vessels and nerves. Inflow rises, outflow is restricted, the tissue fills. That requires working arteries, intact nerve conduction and a vessel lining that responds to the right signals.
Testosterone governs something else: desire above all, along with drive, mood, muscle mass and bone density. It also influences penile tissue and the signalling pathways that trigger an erection, but it is not the engine.
From this follows the most useful distinction in practice: if desire has gone, that points to the hormonal side. If desire is intact and the erection fails anyway, that points to the vascular side.
The difference in everyday life
No single feature proves anything on its own, but the patterns differ clearly:
| Observation | Points more towards |
|---|---|
| Desire has disappeared, not just the erection | hormonal side |
| Lack of drive, low mood, loss of muscle | hormonal side |
| Symptoms developed slowly over months | hormonal side |
| Desire unchanged, the erection fails | vascular side |
| No more morning or nighttime erections | organic cause, either possible |
| It works in some situations and not in others | psychogenic cause |
| Smoking, diabetes, high blood pressure, raised lipids | vascular side |
| Sudden onset after a stressful event | psychogenic cause |
The last point is taken too lightly. An erection problem that comes and goes with the situation, and is absent during masturbation, is rarely hormonal, and a testosterone preparation will not change it. Isolated loss of desire is a different case, covered in low libido in men.
What testosterone actually delivers
Here there are solid numbers rather than promises. A meta-analysis pooled 14 placebo-controlled randomised trials with 2,298 participants, mean age 60, mean follow-up 40 weeks, measured with the standard international questionnaire for erectile function [1].
The result: erectile function improved by 2.31 points on testosterone compared with placebo. What matters is how that figure breaks down:
- In men with a pronounced deficiency below 8 nmol/L the improvement was 2.95 points.
- In men with a milder deficiency below 12 nmol/L it was only 1.47 points.
- With diabetes and obesity the effect was smaller.
Desire, orgasm and overall sexual satisfaction improved as well [1]. That is a real effect, and it grows the lower the starting value. It is not a substitute for treating the vascular side: the authors consider testosterone alone defensible in milder erectile dysfunction but recommend adding further measures where the dysfunction is more severe [1].
Put plainly: a man with a testosterone level in the normal range who hopes for hormone therapy will be disappointed. A man with a clear deficiency can expect a measurable improvement, though not necessarily the solution to the whole problem.
Why an erection problem is a warning sign
This is the section missing from advertising copy, and it is the most important one. The vessels in the penis are narrow. When vessel walls change throughout the body, it shows there earlier than at the heart.
An Australian cohort study followed 95,038 men aged 45 and over, linking questionnaire data with hospital admissions and deaths. Among men without pre-existing cardiovascular disease, severe erectile dysfunction compared with none carried [2]:
| Event | Relative risk |
|---|---|
| Coronary heart disease | 1.60 (1.31–1.95) |
| Heart failure | 8.00 (2.64–24.2) |
| Peripheral arterial disease | 1.92 (1.12–3.29) |
| All cardiovascular disease combined | 1.35 (1.19–1.53) |
| All-cause mortality | 1.93 (1.52–2.44) |
Risk rose stepwise with the severity of the dysfunction. The authors conclude that men with erectile dysfunction should have their cardiovascular risk assessed where this has not already been done [2].
A review puts it this way: erectile dysfunction is not only a consequence of cardiovascular disease but often an early sign of disease still to come. And it is an opportunity to look for comorbidities, improving not just sexual health but men's overall health [3].
How are your testosterone levels?
Our doctor-developed self-test shows you in 4 minutes whether an evaluation could be useful — free and anonymous.
Concretely, that means blood pressure, blood glucose or HbA1c, lipids, smoking status and body weight. A man who presents with an erection problem and leaves with nothing but a prescription has been treated incompletely.
Why the order matters
A hormone level alone does not answer the question. A systematic review examined specifically what benefit measuring hormone levels in erectile dysfunction brings for treatment outcomes, and describes that benefit as uncertain [4]. This does not mean you should not measure. It means measuring alone does not establish the cause.
The sensible order looks like this:
- A history that separates desire from erection. This single question often yields more than any laboratory panel.
- Measure testosterone twice, fasting in the morning, confirming an abnormal value on a second day [5]. What else belongs in the panel is covered in TRT blood work.
- Assess cardiovascular risk, independently of the hormone result [2][3].
- Only then decide whether hormone therapy, treatment of the vascular side, or both makes sense.
If your testosterone is normal
For many men that is the disappointing answer, and it is still a good one: there is a route, it simply runs the other way round.
Treat the vascular side. The standard drug treatment for erectile dysfunction is PDE5 inhibitors. The same systematic review that describes the benefit of hormone testing as uncertain compared them with hormonal approaches [4]. They are prescription-only, well studied, and work independently of the testosterone level.
Address the causes. Stopping smoking, regular exercise, weight reduction and properly controlled blood pressure and glucose all act on the same vascular function an erection depends on. That is slower than a tablet, and it addresses the cause.
Look more closely in younger men. Penile doppler ultrasound can provide important information about cardiovascular risk, particularly in younger men with an otherwise low risk profile [3]. Erectile dysfunction at 40 is a more serious signal than at 70.
Do not dismiss the psychological side. Performance pressure, relationship conflict, depression and some medications, certain antidepressants above all, are common causes. Here a conversation helps more than any hormone.
When both apply
The most common case in practice is not one or the other but the mixture: a man with excess weight, raised blood glucose, low testosterone and an erection problem. Here the measures work together.
Testosterone improves desire and erectile function measurably, especially at low baseline values [1]. The effect is smaller with diabetes and obesity, which is not an argument against therapy but one for treating the excess weight alongside it.
If you are unsure which of the two sides dominates in your case, the online self-test gives a first indication. Making the distinction belongs in a conversation, because it depends on your account rather than on a laboratory value alone, and that is easily arranged by telemedicine. Which complaints point to a deficiency is set out in testosterone deficiency symptoms.
FAQ
Is erectile dysfunction a sign of low testosterone?
Sometimes, but not usually. What points to the hormonal side is above all lost desire, often together with low drive and low mood. If desire remains and only the erection fails, the cause more often lies in vessels, nerves or the mind.
Does testosterone help erectile dysfunction?
Measurably, where a deficiency is documented. In a meta-analysis of 14 randomised trials with 2,298 men, erectile function improved by 2.31 points over placebo, by 2.95 points with a pronounced deficiency and by 1.47 with a milder one [1]. With a normal testosterone level, no effect should be expected.
Why should I have a heart check when this is about sex?
Because erectile dysfunction is often the first sign of vascular disease. Among 95,038 men, severe erectile dysfunction was associated with a 60 percent higher risk of coronary heart disease and nearly double the all-cause mortality [2]. That assessment is the more valuable part of the visit.
I have diabetes. Will hormone therapy do anything for me?
Yes, though less strongly. The effect on erectile function is smaller in men with diabetes and obesity than in men without them [1]. That argues for treating the metabolic situation in parallel, not against therapy.
It works sometimes and not others. What does that mean?
That argues against a hormonal or purely physical cause. When erections occur without difficulty in some situations and fail in others, the emphasis usually lies in the psychological and relational domain. Hormone therapy is not the right route then.
Further Reading

Specialist in General Internal Medicine · treating physician
This article was medically reviewed by Dr. Ramadan for accuracy. It is based on current research and international guidelines.
Your Next Step
You've informed yourself — now you can find out in 4 minutes whether a medical evaluation makes sense for you.
Sources
- [1]Corona G et al. "Meta-analysis of Results of Testosterone Therapy on Sexual Function Based on International Index of Erectile Function Scores." Eur Urol. 2017;72(6):1000-1011. PubMed
- [2]Banks E et al. "Erectile dysfunction severity as a risk marker for cardiovascular disease hospitalisation and all-cause mortality: a prospective cohort study." PLoS Med. 2013;10(1):e1001372. PubMed
- [3]Corona G et al. "Erectile dysfunction and cardiovascular risk: a review of current findings." Expert Rev Cardiovasc Ther. 2020;18(3):155-164. PubMed
