Skip navigation

Erectile Dysfunction Treatments


There is now a major set of stories on erectile dysfunction and erectile dysfunction treatments in Bandolier's Sexual Health pages.

Erectile dysfunction has been defined by a National Institutes of Health conference as the inability to achieve or maintain an erection sufficient for satisfactory sexual performance. Erectile dysfunction is strongly age related. As many as 30 million men may be affected in the USA [1]. While there is an estimated prevalence across all ages of about 10% (making erectile dysfunction common), the prevalence rises to over 50% in men between 50 and 70 years of age, though it is not an inevitable consequence of normal ageing [1]. It is also associated with a number of organic disorders and diseases. In diabetes, for instance, it occurs in up to 40% of men. But erectile dysfunction may also occur with cardiovascular disorders (especially in men with angina or after myocardial infarction), neurological disorders, after pelvic surgery or trauma, and as a consequence of pharmacological treatments of a number of diseases [2].

Conditions that are common often become expensive if an effective treatment comes along. For erectile dysfunction what most of us know (or think) is that treatments are reserved for funny boxes in the pages of our more popular newspapers. We might be dimly aware that a number of physical treatments have been used, including semi-rigid or inflatable implants, vacuum constriction devices, and vascular surgery. Some of these can be effective in some men, but they can be expensive and may not be appropriate in every affected man.

New treatments

Bandolier has noticed that new pharmacological treatments are being reported in large and well-conducted trials in our top medical journals. Alprostidil (prostaglandin E1) relaxes corpora cavernosal smooth muscle in vitro, and is effective in the treatment of impotence. Another treatment is sildenafil, which is a phosphodiesterase inhibitor, which also relaxes cavernosal smooth muscle.

The new treatments are effective but delivery systems have been complicated. The idea of intracavernosal injections brings tears to the eyes of many. Intraurethral application seems better, but a pill? Now you're talking.

Just imagine for a moment that there was an effective (and safe) oral treatment that gave men with erectile dysfunction (and their partners) what they wanted. The demand would be large - 10% of all men is 2.5 million in the UK. Half of all men between 50 and 70 (say 300,000 in each year) translates into 3 million men. And all those tired executives.....?

So while treatment of erectile dysfunction may be minor league stuff at the moment, it has the potential to make huge demands on health services. Commissioners need to begin to think about this now.

Intracavernosal alprostadil

A randomised, double-blind, placebo-controlled study [3] investigated different doses of alprostadil injected in to the base of the penis by a nurse. Men in the trial had at least a four month history of erectile dysfunction and were in stable, monogamous heterosexual relationships.

There was a dose-response relationship with intracavernosal alprostadil, with number-needed-to-treat of about 2 at higher doses to produce an erection sufficient for intercourse. The mean duration of erection was also dose-related, being about 12 minutes with 2.5 µg but over 30 minutes for the higher doses. Priapism developed in one man, and two men had erections which lasted for more than four hours.

Intracavernosal alprostadil µg Number needed to treat (95% CI)
2.5 5.9 (3.7 - 14)
5 3.6 (2.5 - 6.0)
10 2.2 (1.7 - 3.1)
20 2.0 (1.6 - 2.7)

An open-label part of the study, of 683 men who used intracavernosal injections at home over a six month period, demonstrated that this method could be used successfully. Satisfactory sexual activity was reported by men as occurring after 87% of 13,762 injections, and by their partners after 86% of 9,892 injections (not all of them filled out their forms).

Transurethral alprostadil

Men with chronic erectile difficulties and their partners participated in a multi-centre study in the USA [4]. Men had to be in stable, monogamous, heterosexual relationships. All the men (1511) had to have an erectile dysfunction of at least three months duration with a primary organic cause. The study was in two phases. Phase 1 was an evaluation in the clinic of escalating doses of alprostadil using a device which allowed a semi-solid pellet to be applied some 3 cm into the urethra. This was done immediately after urination so that any residual urine would work both to facilitate introduction of the 3.5 mm diameter device, and dissolution of the pellet. Men began with a 125 µg dose or 250 µg, chosen at random; they then used the other dose on a subsequent occasion. Doses were identical in appearance and the administration was double-blind as well as random. Those men who did not have a satisfactory response with 250 µg went on to use 500 µg and 1000 µg doses, again in a random, double blind evaluation of these two doses. A satisfactory response was graded with an erection assessment scale according to the following score:

  1. no response
  2. some enlargement
  3. full enlargement but insufficient rigidity
  4. erection sufficient for intercourse
  5. full rigidity

Scores were assigned by the men (and confirmed by the investigator), and scores of 4 or 5 were deemed a success. Men also scored their erection as comfortable or uncomfortable. In this phase, 996 men had an erection (score 4 or 5). There was a clear dose-response for effect, but the proportion of men who found it uncomfortable did not differ. Very few men rated their erection as very uncomfortable.

For all effective doses, the average time to onset of response was about seven minutes, with a maximal response in about 20 minutes, with the erection lasting for about 30 to 40 minutes. There was no effect of age or cause of dysfunction.

Phase 2 was a randomised, double-blind, placebo-controlled evaluation of alprostadil at home in the 996 men who achieved success in the clinic. This has to be thought of as an enriched enrolment. The choice of dose for the men was that which had worked successfully and comfortably for them in the clinic. Successful outcomes were the achievement of at least one intercourse and/or orgasm in a three month period, and any adverse effects were noted.

The main results, expressed as relative benefit or risk and number-needed-to-treat or to harm on an intention-to-treat analysis, are shown below. One in two to three men treated with transurethral alprostadil achieved an erection which enabled them to have intercourse and achieve an organism who would not have done had they received placebo. This was at a cost of some mild penile pain for one in three, mild urethral trauma in one in 25 and dizziness in one in 50.

Outcome Number with outcome on alprostadil Number with outcome on placebo Relative benefit or risk (95% CI) Number needed to treat or harm (95%CI)
Intercourse 299/486 93/511 3.4 (2.8 - 4.1) 2.3 (2.1 - 2.6)
Orgasm 293/486 118/511 2.6 (2.2 - 3.1) 2.7 (2.3 - 3.2)
Pain 159/486 17/511 9.8 (6.1 - 16) 3.5 (3.0 - 4.0)
Trauma 35/486 5/511 5.3 (2.0 - 14) 25 (16 - 49)
Dizziness 9/486 1/511 9.5 (1.2 - 74) 50 (34 - 250)

Adverse effects with alprostadil

Adverse effects were higher after intracavernosal injection than after transurethral application. This was mainly because of a higher number of men reporting penile pain, although penile pain did not occur after a every injection or application. Prolonged erection and priapism did not occur with transurethral application, but this was a problem with intracavernosal injection.

Oral sildenafil

We have only one randomised trial for sildenafil on only 12 men [5]. This is a slightly difficult paper to follow, but again was done in two phases, both of high methodological quality. It was conducted in Bristol on 12 men with more than one and half years history of erectile dysfunction. The first phase was in the laboratory and was aimed at dose-finding. A second randomised cross-over seven day phase at home yielded a mean of 6 erections (95% CI 3 to 11) over seven days for men receiving sildenafil, compared with 1.3 (0.5 to 3) for placebo. On a crude global scale, 10 of 12 men reported improved erectile activity with sildenafil compared with 2 out of 10 with placebo.

Bandolier has also seen an abstract of a larger study [6] on 351 men, implying an NNT of 2 for the highest dose of 50 mg sildenafil where 89% of men reported that sildenafil improved their erections.


Not all of the treatments are yet licensed in the UK. Perhaps some of them may never be, and if they are, it will be some years away. But there are problems right now. How would you advise a patient who wanted a one week prescription for seven injections of alprostadil at £10 a time? There are hard decisions to be taken.


  1. HA Feldman, I Goldstein, DG Hatzichristou et al. Impotence and its medical and psychosocial correlates: results of the Massechusetts Male Ageing Study. Journal of Urology 1994 151: 54-61.
  2. KE Hawton. Sexual problems associated with physical illness. Oxford Textbook of Medicine (eds DJ Weatherall, JG Ledingham, DA Warrell), Oxford University Press, 3rd edition, 1997: 4243-7.
  3. OI Linet, FG Ogrinc. Efficacy and safety of intracavernosal alprostadil in men with erectile dysfunction. New England Journal of Medicine 1996 334: 873-7.
  4. H Padma-Nathan, WJG Hellstrom, FE Kaiser et al. Treatment of men with erectile dysfunction with transurethral alprostadil. New England Journal of Medicine 1997 336: 1-7.
  5. M Boolell, S Gepi-Attee, JC Gingell, MJ Allen. Sildenafil, a novel effective oral therapy for male erectile dysfunction. British Journal of Urology 1996 78: 257-261.
  6. CJ Gingell, A Jardin, AM Olsen et al. A new oral treatment for erectile dysfunction: a double-blind, placebo-controlled, once daily dose response study. Proceedings of the American Urological Association, 1966, 155, 495A.

previous or next story in this issue