Skip to main content
NCBI home page
BMJ Clinical Evidence logoLink to BMJ Clinical Evidence
. 2006 Mar 1;2006:1803.

Erectile dysfunction

Prathap Tharyan 1,#, Ganesh Gopalakrishanan 2,#
PMCID: PMC2907627

Abstract

Introduction

Erectile dysfunction may affect 30-50% of men aged 40-70 years, with age, smoking and obesity being the main risk factors, although 20% of cases have psychological causes.

Methods and outcomes

We conducted a systematic review and aimed to answer the following clinical questions: What are the effects of phosphodiesterase inhibitors in men with erectile dysfunction of any cause? What are the effects of phosphodiesterase inhibitors on erectile dysfunction in men with diabetes, with cardiovascular disease, with spinal cord injury, and with prostate cancer or prostatectomy? What are the effects of drug treatments other than phosphodiesterase inhibitors in men with erectile dysfunction of any cause? What are the effects of devices, psychological/behavioural treatments, and alternative treatments in men with erectile dysfunction of any cause? We searched: Medline, Embase, The Cochrane Library and other important databases up to August 2005 (BMJ Clinical Evidence reviews are updated periodically, please check our website for the most up-to-date version of this review). We included harms alerts from relevant organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA).

Results

We found 63 systematic reviews, RCTs or observational studies that met our inclusion criteria. We performed a GRADE evaluation of the quality of evidence for interventions.

Conclusions

In this systematic review we present information relating to the effectiveness and safety of the following interventions: alprostadil (intracavernosal, intraurethral, topical) apomorphine, cognitive behavioural therapy, ginseng, papaverine, papaverine plus phentolamine (bimix), papaverine plus phentolamine plus alprostadil (trimix), penile prostheses, phosphodiesterase inhibitors (sildenafil, tadalafil, vardenafil), psychosexual counselling, vacuum devices, and yohimbine.

Key Points

Erectile dysfunction may affect 30-50% of men aged 40-70 years, with age, smoking and obesity being the main risk factors, although 20% of cases have psychological causes.

Sildenafil improves erections and increases the likelihood of successful intercourse overall and in men with diabetes mellitus, heart disease, spinal cord injury, prostate cancer or after radical prostatectomy.

  • Tadalafil and vardenafil are also effective overall and in men with diabetes. Vardenafil may be effective after prostatectomy, but few studies have been found.

CAUTION: sildenafil, tadalafil and vardenafil are contraindicated in men who are taking nitrates as combined treatment has been associated with severe hypotension and death.

Intracavernosal alprostadil improves erections compared with placebo, intraurethral alprostadil and intracavernosal papaverine, but can cause penile pain in up to 40% of men.

  • Intracavernosal alprostadil may be as effective as sildenafil and bimix, while topical alprostadil may also be effective.

  • Adding phentoloamine to intracavernosal papaverine (bimix) may increase effectiveness compared with papaverine alone, and adding alprostadil to bimix (trimix) may be more effective again. However, papaverine injections may cause altered liver function, and penile bruising and fibrosis.

  • Sublingual apomorphine, ginseng and yohimbine may increase successful erections and intercourse compared with placebo.

Vacuum devices may be as effective as intracavernosal alprostadil at increasing rigidity, but less effective for orgasm, and may block ejaculation.

  • There is consensus that penile prostheses may be beneficial, but they can cause infections and are only used if less invasive treatments have failed.

Psychosexual counselling and cognitive behavioural therapy may improve sexual functioning in men with psychological erectile dysfunction, but few good quality studies have been found.

About this condition

Definition

Erectile dysfunction is defined as the persistent inability to obtain or maintain sufficient rigidity of the penis to allow satisfactory sexual performance. The term erectile dysfunction has largely replaced the term "impotence". For the purposes of this review we included only men with normal testosterone and gonadotrophin levels, who could gain an erection while asleep. We also included men with comorbid conditions such as cardiovascular disorders, prostate cancer, diabetes, and spinal cord injury. We excluded men with drug induced sexual dysfunction. Because the cause of erectile dysfunction in men with cardiovascular disease is unclear (the disease or treatment drugs), we included them.

Incidence/ Prevalence

Cross-sectional epidemiological studies from around the world reveal that 30-50% of men aged 40-70 years report some degree of erectile dysfunction. About 150 million men worldwide are unable to achieve and maintain an erection adequate for satisfactory sexual intercourse. Age is the variable most strongly associated with erectile dysfunction; between the ages of 40 to 70 years, the incidence of moderate erectile dysfunction doubles from 17% to 34%, whereas that of severe erectile dysfunction triples from 5% to 15%.

Aetiology/ Risk factors

About 80% of cases are believed to have an organic cause, the rest being psychogenic in origin. Most cases of erectile dysfunction are believed to be multifactorial and secondary to disease, stress, trauma (such as spinal cord injury, pelvic and prostate surgery), or drug adverse effects that interfere with the coordinated psychological, neurological, endocrine, vascular, and muscular factors necessary for normal erections. Risk factors include increasing age, smoking, and obesity. The prevalence of erectile dysfunction also increases in people with diabetes mellitus, hypertension, heart disease, anxiety, and depression.

Prognosis

We found no good evidence on prognosis in untreated organic erectile dysfunction.

Aims of intervention

To restore satisfactory erections, with minimal adverse effects of treatment.

Outcomes

Self and partner reports of satisfaction and sexual function, quality of life, objective tests of penile rigidity, time to take effect, duration of effect, ease of usage, mortality, and adverse effects of treatment.

Methods

Clinical Evidence search and appraisal August 2005. We searched for RCTs comparing each listed intervention versus placebo, no treatment, or each other and have included all studies of sufficient quality. We excluded studies where less than 80% of men had normal hormone levels. We only included studies that considered clinical end points. Where several RCTs were found, we have only included studies that used clinical end points (such as successful attempts at intercourse) as outcome measures. We have performed a GRADE evaluation of the quality of evidence for interventions included in this review (see table ).

Table.

GRADE evaluation of interventions for erectile dysfunction

Important outcomes Improvement in sexual function, adverse effects
Number of studies (participants) Outcome Comparison Type of evidence Quality Consistency Directness Effect size GRADE Comment
What are the effects of phosphodiesterase inhibitors in men with erectile dysfunction of any cause?
29 (at least 2283 men) Improvement in sexual function Sildenafil v placebo in men with erectile dysfunction of any cause 4 −1 0 0 0 Moderate Quality point deducted for for uncertainty about methods of randomisation and allocation concealment
1 (54) Improvement in sexual function Sildenafil v intracavernosal alprostadil in men with erectile dysfunction of any cause 4 −1 0 0 0 Moderate Quality point deducted for sparse data
16 (2827) Improvement in sexual function Tadalafil v placebo in men with erectile dysfunction of any cause 4 −2 0 −2 0 Very low Quality points deducted for uncertainty about methods of randomisation and whether allocation concealmentwas performed, and for lack of homogeneity in study design and outcome assessments. Directness points deducted for heterogeneity in aetiology of erectile dysfunction and age of population
3 (1729) Improvement in sexual function Vardenafil v placebo in menwith erectile dysfunction of any cause 4 −1 0 −1 0 Low Quality point deducted for uncertainty about quality of studies. Directness point deducted for inclusion of previous responders to treatment
What are the effects of phosphodiesterase inhibitors on erectile dysfunction in men with diabetes?
18 (1820) Improvement in sexual function Sildenafil v placebo in men with diabetes 4 −1 0 0 +1 High Quality point deducted for subgroup analysis. Effect size point added for RR greater than 2 but less than 5
1 (216) Improvement in sexual function Tadalafil v placebo in men with diabetes 4 0 0 0 0 High
1 (452) Improvement in sexual function Vardenafil v placebo in men with diabetes 4 0 0 0 0 High
What are the effects of phosphodiesterase inhibitors on erectile dysfunction in men with cardiovascular disease?
At least 2 RCTs (739) Improvement in sexual function Sildenafil v placebo in men with heart disease 4 0 0 0 0 High
24 (664) Adverse effects Sildenafil v placebo in men with heart disease 4 0 0 0 0 High
What are the effects of phosphodiesterase inhibitors on erectile dysfunction in men with spinal cord injury?
1 (205) Improvement in sexual function Sildenafil v placebo in men with spinal cord injury 4 0 0 0 +2 High Effect size points added for RR greater than 5
What are the effects of phosphodiesterase inhibitors on erectile dysfunction in men with prostate cancer or prostatectomy?
2 (176) Improvement in sexual function Sildenafil v placebo in men after radical prostatectomy or prostate cancer 4 −2 0 0 +1 Moderate Quality points deducted for sparse data and for subgroup analysis. Effect size point added for RR less than 5 but greater than 2
1 (440) Improvement in sexual function Vardenafil v placebo in men after prostatectomy 4 −1 0 −1 0 Low Quality point deducted for for uncertainty about quality of studies. Directness point deducted for inclusion of previous responders to treatment
What are the effects of drug treatments other than phosphodiesterase inhibitors in men with erectile dysfunction of any cause?
3 (1828) Improvement in sexual function Intraurethral alprostadil v placebo in men with erectile dysfunction of any cause 4 −1 0 −1 +2 High Quality point deducted for for not stating methods of randomisation and whether allocation concealment was performed. Directness point deducted for pre-selecting treatment responders affecting generalisability to clinical practice. Effect size points added for OR greater than 5
1 (270) Improvement in sexual function Intraurethral alprostadil v placebo in men after radical prostatectomy 4 −1 0 0 0 Moderate Quality point deducted for not reporting method of randomisation and allocation concealment
3 (274) Improvement in sexual function Intraurethral alprostadil v intracavernosal alprostadil in men with erectile dysfunction of any cause 4 −2 0 −1 0 Very low Quality points deducted for open label study and for uncertaintity about randomisation and whether allocation concealment was performed . Directness point deducted for pre-selecting treatment responders affecting generalisability to clinical practice
3 (1345) Improvement in sexual function Apomorphine v placebo in men with erectile dysfunction of any cause 4 −1 0 0 0 Moderate Quality point deducted for not reporting methods of randomisation and allocation concealment
2 (82) Improvement in sexual function Topical alprostadil v placebo in men with erectile dysfunction of any cause 4 −2 0 0 0 Low Quality points deducted for sparse data, and for not reporting methods of randomisation and allocation concealment
1 (40) Improvement in sexual function Papaverine v papaverine plus phentolamine (bimix) in men with with erectile dysfunction of any cause 4 −1 0 0 0 Moderate Quality point deducted for sparse data
1 (30) Improvement in sexual function Papaverine plus phentolamine (bimix) v placebo in men with with erectile dysfunction of any cause 4 −1 0 −1 0 Low Quality point deducted for sparse data. Directness point deducted for no direct statistical comparison between groups
2 (356) Improvement in sexual function Intracavernosal alprostadil v placebo in men with erectile dysfunction of any cause 4 −3 0 0 0 Very low Quality points deducted for incomplete reporting of results, and for methodological weaknesses (randomisation/allocation concealment, subjective assessment of outcome and unblinded assessment of outcome)
3 (235) Improvement in sexual function Intracavernosal alprostadil v papaverine in men with erectile dysfunction of any cause 4 −3 0 0 0 Very low Quality points deducted for methodological weaknesses (uncertainty about methods of randomisation and allocation concealment, subjective and unblinded assessment of outcome)
2 (142) Improvement in sexual function Intracavernosal alprostadil v papaverine plus phentolamine (bimix) in men with erectile dysfunction of any cause 4 −3 0 0 0 Very low Quality points deducted for sparse data, and for methodological weaknesses (uncertainty about methods of randomisation and allocation concealment, subjective and unblinded assessment of outcome)
2 (114) Improvement in sexual function Intracavernosal alprostadil v alprostadil plus papaverine plus phentolamine (trimix) in men with erectile dysfunction of any cause 4 −3 0 0 0 Very low Quality points deducted for sparse data, and for methodological weaknesses (uncertainty about methods of randomisation and allocation concealment, subjective and unblinded assessment of outcome)
1 (50) Improvement in sexual function Intracavernosal alprostadil v vacuum devices in men with erectile dysfunction of any cause 4 −3 0 −1 0 Very low Quality points deducted for sparse data, and for methodological weaknesses (randomisation/allocation concealment, subjective assessment of outcome). Directness point deducted for not using validated outcome assessments
What are the effects of devices in men with erectile dysfunction of any cause?
1 (45) Improvement in sexual function Vacuum devices plus psychosexual therapy v psychosexual therapy in men with erectile dysfunction of any cause 4 −3 0 −1 0 Very low Quality points deducted for sparse data, and for methodological weaknesses (lack of blinding, subjective assessment of outcome). Directness point deducted for not using validated outcome assessments
What are the effects of psychological/behavioural treatments in men with erectile dysfunction of any cause?
2 (90) Improvement in sexual function Psychosexual counselling v waiting list control in men with erectile dysfunction of any cause 4 −3 0 0 0 Very low Quality points deducted for sparse data, incomplete reporting of results and for methodological weaknesses (uncertainities about methods of randomisaton and /allocation concealment
1 (69) Improvement in sexual function Psychosexual counselling vinterpersonal therapy in men with erectile dysfunction of any cause 4 −2 0 0 0 Low Quality points deducted for sparse data, and for methodological weaknesses (uncertainities about methods of randomisaton and /allocation concealment
What are the effects of alternative treatments in men with erectile dysfunction of any cause?
1 (45) Improvement in sexual function Ginseng v placebo in men with erectile dysfunction of any cause 4 −1 0 0 0 Moderate Quality point deducted for sparse data
8 (448) Improvement in sexual function Yohimbine v placebo in men with with erectile dysfunction of any cause 4 −3 0 −2 +1 Very low Quality points deducted for methodological weaknesses (uncertainity about method of randomisation, lack of homogeneity in study design and outcome assessments). Directness points deducted for heterogeneity in aetiology of erectile dysfunction and age of population, and for method of assessment
Open in a new tab

Type of evidence: 4 = RCT; 2 = Observational Consistency: similarity of results across studies Directness: generalisability of population or outcomes Effect size: based on relative risk or odds ratio

Glossary

Ecchymosis

The skin discoloration caused by the escape of blood into the tissues from ruptured blood vessels.

Global Assessment Questionnaire

A self administered questionnaire that allows men to rate improvement in erectile function.

Global Efficacy Question

Asks, “Did the treatment you have been taking over the past 4 weeks improve your erections?” This question is answered with a “yes” or “no”. In some trials, responses are scored on a 7 point scale ranging from “no improvement” to “intense improvement”.

High-quality evidence

Further research is very unlikely to change our confidence in the estimate of effect.

International Index of Erectile Function (IIEF) questions 3 and 4

The questions have been validated for assessing the effects of treatments for men with erectile dysfunction. They ask, “over the past 4 weeks, when you have attempted sexual intercourse, how often were you able to penetrate (enter) your partner?”, and “Over the past 4 weeks, during sexual intercourse, how often were you able to maintain your erection after you have penetrated (entered) your partner?” Each question is answered on a 6 point scale of 0–5.

Low-quality evidence

Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.

Moderate-quality evidence

Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.

Priapism

Prolonged, and often painful, erections of the penis in the absence of sexual desire that can last for several hours to days. Prompt treatment to relieve the erection and prevent scarring is recommended if the erection does not subside in 4 hours.

Sexual Encounter Profile (SEP) questions 2 and 3

This is a diary maintained by men after each sexual attempt consisting of a series of yes/no questions regarding specific aspects of each encounter. Question 2 asks, “Were you able to insert your penis into your partner's vagina?” and question 3 asks, “Did your erection last long enough for you to complete intercourse with ejaculation?”

Very low-quality evidence

Any estimate of effect is very uncertain.

Disclaimer

The information contained in this publication is intended for medical professionals. Categories presented in Clinical Evidence indicate a judgement about the strength of the evidence available to our contributors prior to publication and the relevant importance of benefit and harms. We rely on our contributors to confirm the accuracy of the information presented and to adhere to describe accepted practices. Readers should be aware that professionals in the field may have different opinions. Because of this and regular advances in medical research we strongly recommend that readers' independently verify specified treatments and drugs including manufacturers' guidance. Also, the categories do not indicate whether a particular treatment is generally appropriate or whether it is suitable for a particular individual. Ultimately it is the readers' responsibility to make their own professional judgements, so to appropriately advise and treat their patients.To the fullest extent permitted by law, BMJ Publishing Group Limited and its editors are not responsible for any losses, injury or damage caused to any person or property (including under contract, by negligence, products liability or otherwise) whether they be direct or indirect, special, incidental or consequential, resulting from the application of the information in this publication.

Contributor Information

Prathap Tharyan, Department Of Psychiatry, Christian Medical College, Tamil Nadu, India.

Ganesh Gopalakrishanan, Department Of Psychiatry, Christian Medical College, Tamil Nadu, India.

References

  • 1.McKinley JB. The worldwide prevalence and epidemiology of erectile dysfunction. Int J Impot Res 2000;12:S6–S11. [DOI] [PubMed] [Google Scholar]
  • 2.Mak R, De Backer G, Kornitzer M, at al. Prevalence and correlates of erectile dysfunction in a population-based study in Belgium. Eur Urol 2002;41:132–138. [DOI] [PubMed] [Google Scholar]
  • 3.Moreira ED, Lisboa LCF, Glasser DB. A cross-sectional population based study of erectile dysfunction epidemiology in northeastern Brazil. J Urol 2000;163(suppl):15. [Google Scholar]
  • 4.Feldman HA, Goldstein I, Hatzichristou DG, et al. Impotence and its medical and psychosocial correlates: results of the Massachusetts Male Aging Study. J Urol 1994;151:54–61. [DOI] [PubMed] [Google Scholar]
  • 5.AACE Male Sexual Dysfunction Task Force. American Association of Clinical Endocrinologists medical guidelines for clinical practice for the evaluation and treatment of male sexual dysfunction: a couple's problem – 2003 update. Endocr Pract 2003;9:77–95. [DOI] [PubMed] [Google Scholar]
  • 6.Fink HA, MacDonald R, Rutks IR, et al. Sildenafil for male erectile dysfunction: a systematic review and meta-analysis. Arch Intern Med 2002;162:1349–1360. Search date 2000; primary sources Medline, Healthstar, Current Contents, The Cochrane Library, bibliographies of retrieved trials and review articles, and hand searches of urology journals and national meeting abstracts. [DOI] [PubMed] [Google Scholar]
  • 7.Shirai M, Tsukamoto T, Sato Y, et al. Randomised, placebo-controlled, double-blind study of oral sildenafil in Japanese men with erectile dysfunction. Nishinihon J Urol 2000;62:373–382. [In Japanese] [Google Scholar]
  • 8.Chen KK, Hsieh JT, Huang ST, et al. ASSESS-3: a randomised, double-blind, flexible-dose clinical trial of the efficacy and safety of oral sildenafil in the treatment of men with erectile dysfunction in Taiwan. Int J Impot Res 2001;13:221–229. [DOI] [PubMed] [Google Scholar]
  • 9.Lewis R, Bennett CJ, Borkon WD, et al. Patient and partner satisfaction with Viagra (sildenafil citrate) treatment as determined by the Erectile Dysfunction Inventory of Treatment Satisfaction Questionnaire. Urology 2001;57:960–965. [DOI] [PubMed] [Google Scholar]
  • 10.Meuleman E, Cuzin B, Opsomer RJ, et al. A dose-escalation study to assess the efficacy and safety of sildenafil citrate in men with erectile dysfunction. BJU Int 2001;87:75–81. [DOI] [PubMed] [Google Scholar]
  • 11.Seidman SN, Roose SP, Menza MA, et al. Treatment of erectile dysfunction in men with depressive symptoms: results of a placebo-controlled trial with sildenafil citrate. Am J Psychiatry 2001;158:1623–1630. [DOI] [PubMed] [Google Scholar]
  • 12.Becher E, Tejada Noriega A, Gomez R, et al. Sildenafil citrate (Viagra) in the treatment of men with erectile dysfunction in southern Latin America: a double-blind, randomized, placebo-controlled, parallel-group, multicenter, flexible-dose escalation study. Int J Impot Res 2002;14(suppl):S33–S41. [DOI] [PubMed] [Google Scholar]
  • 13.Glina S, Bertero E, Claro J, et al. Efficacy and safety of flexible-dose oral sildenafil citrate (Viagra) in the treatment of erectile dysfunction in Brazilian and Mexican men. Int J Impot Res 2002;14(suppl):27–32. [DOI] [PubMed] [Google Scholar]
  • 14.Gomez F, Davila H, Costa A, et al. Efficacy and safety of oral sildenafil citrate (Viagra) in the treatment of male erectile dysfunction in Colombia, Ecuador, and Venezuela: a double-blind, multicenter, placebo-controlled study. Int J Impot Res 2002;14(suppl):42-47. [DOI] [PubMed] [Google Scholar]
  • 15.Young JM, Bennett C, Gilhooly P, et al. Efficacy and safety of sildenafil citrate (Viagra) in black and Hispanic American men. Urology 2002;60:39–48. [DOI] [PubMed] [Google Scholar]
  • 16.Choi HK, Ahn TY, Kim JJ, et al. A double-blind, randomised, placebo, controlled, parallel group, multicentre, flexible-dose escalation study to assess the efficacy and safety of sildenafil administered as required to male outpatients with erectile dysfunction in Korea. Int J Impot Res 2003;15:80–86. [DOI] [PubMed] [Google Scholar]
  • 17.Kongkanand A, Ratana-Olarn K, Ruangdilokrat S, et al. The efficacy and safety of oral sildenafil in Thai men with erectile dysfunction: a randomized, double-blind, placebo controlled, flexible-dose study. J Med Assoc Thailand 2003;86:195–205. [PubMed] [Google Scholar]
  • 18.Levinson IP, Khalaf IM, Shaeer KZ, et al. Efficacy and safety of sildenafil citrate (Viagra) for the treatment of erectile dysfunction in men in Egypt and South Africa. Int J Impot Res 2003;15(suppl):S25–S29. [DOI] [PubMed] [Google Scholar]
  • 19.Nurnberg HG, Hensley PL, Gelenberg AJ, et al. Treatment of antidepressant-associated sexual dysfunction with sildenafil. JAMA 2003;289:56–64. [DOI] [PubMed] [Google Scholar]
  • 20.Padma-Nathan H, Stecher VJ, Sweeney M, et al. Minimal time to successful intercourse after sildenafil citrate: results of a randomized, double-blind, placebo-controlled trial. Urology 2003;62:400–403. [DOI] [PubMed] [Google Scholar]
  • 21.Tignol J, Furlan PM, Gomez-Beneyto M, et al. Efficacy of sildenafil citrate (Viagra) for the treatment of erectile dysfunction in men in remission from depression. Int Clin Psychopharmacol 2004;19:191–199. [DOI] [PubMed] [Google Scholar]
  • 22.Ethans KD, Casey AR, Schryvers OI, et al. The effects of sildenafil on the cardiovascular response in men with spinal cord injury at or above the sixth thoracic level. J Spinal Cord Med 2003;26:222–226. [DOI] [PubMed] [Google Scholar]
  • 23.Wang ZL, Li B, Yan JZ, et al. Prostaglandin E1 versus sildenafil in the management of erectile dysfunction. Zhonghua Nan Ke Xue 2002;8:198–200. [In Chinese] [PubMed] [Google Scholar]
  • 24.Carson CC, Rajfer J, Eardley I, et al. The efficacy and safety of tadalafil: an update. BJU Int 2004;93:1276–1281. [DOI] [PubMed] [Google Scholar]
  • 25.Padma-Nathan H, McMurray JG, Pullman WE, et al. On-demand IC351 (Cialis) enhances erectile function in patients with erectile dysfunction. Int J Impot Res 2001;13:2–9. [DOI] [PubMed] [Google Scholar]
  • 26.Seftel AD, Wilson SK, Knapp PM, et al. The efficacy and safety of tadalafil in United States and Puerto Rican men with erectile dysfunction. J Urol 2004;172:652-657. [Erratum in: J Urol 2005;173:664] [DOI] [PubMed] [Google Scholar]
  • 27.Porst H, Padma-Nathan H, Giuliano F, et al. Efficacy of tadalafil for the treatment of erectile dysfunction at 24 and 36 hours after dosing: a randomized controlled trial. Urology 2003;62:121–125. [DOI] [PubMed] [Google Scholar]
  • 28.Govier F, Potempa AJ, Kaufman J, et al. A multicenter, randomized, double-blind, crossover study of patient preference for tadalafil 20 mg or sildenafil citrate 50 mg during initiation of treatment for erectile dysfunction. Clin Ther 2003;25:2709–2723. [DOI] [PubMed] [Google Scholar]
  • 29.von Keitz A, Rajfer J, Segal S, et al. A multicenter, randomized, double-blind, crossover study to evaluate patient preference between tadalafil and sildenafil. Eur Urol 2004;45:499–507. [DOI] [PubMed] [Google Scholar]
  • 30.Crowe SM, Streetman DS. Vardenafil treatment of erectile dysfunction. Ann Pharmacother 2004;38:77–85. Search date 2002; primary sources Medline and bibliographic searches of retrieved articles, unpublished data from manufacturer, and proceedings of professional meetings. [DOI] [PubMed] [Google Scholar]
  • 31.Hatzichristou D, Montorsi F, Buvat J, et al. The efficacy and safety of flexible-dose vardenafil (Levitra) in a broad population of European men. Eur Urol 2004;45:634–641. [DOI] [PubMed] [Google Scholar]
  • 32.Boulton AJ, Selam JL, Sweeney M, et al. Sildenafil citrate for the treatment of erectile dysfunction in men with Type II diabetes mellitus. Diabetologia 2001;44:1296–1301. [DOI] [PubMed] [Google Scholar]
  • 33.Stuckey BG, Jadzinsky MN, Murphy LJ, et al. Sildenafil citrate for treatment of erectile dysfunction in men with type 1 diabetes: results of a randomized controlled trial. Diabetes Care 2003;26:279–284. [DOI] [PubMed] [Google Scholar]
  • 34.Escobar-Jimenez F. Efficacy and safety of sildenafil in men with type 2 diabetes mellitus and erectile dysfunction. Med Clin (Barc) 2002;119:121–124. [In Spanish] [DOI] [PubMed] [Google Scholar]
  • 35.Safarinejad MR. Oral sildenafil in the treatment of erectile dysfunction in diabetic men: a randomized double-blind and placebo-controlled study. J Diabetes Complications 2004;18:205–210. [DOI] [PubMed] [Google Scholar]
  • 36.Incrocci L, Hop WC, Slob AK. Efficacy of sildenafil in an open-label study as a continuation of a double-blind study in the treatment of erectile dysfunction after radiotherapy for prostate cancer. Urology 2003;62:116–120. [DOI] [PubMed] [Google Scholar]
  • 37.Saenz de Tejada I, Anglin G, Knight JR, et al. Effects of tadalafil on erectile dysfunction in men with diabetes. Diabetes Care 2002;25:2159–2164. [DOI] [PubMed] [Google Scholar]
  • 38.Olsson AM, Persson CA. Efficacy and safety of sildenafil citrate for the treatment of erectile dysfunction in men with cardiovascular disease. Int J Clin Pract 2001;55:171–176. [PubMed] [Google Scholar]
  • 39.DeBusk RF, Pepine CJ, Glasser DB, et al. Efficacy and safety of sildenafil citrate in men with erectile dysfunction and stable coronary artery disease. Am J Cardiol 2004;93:147–153. [Erratum in: Am J Cardiol 2004;94:543–544] [DOI] [PubMed] [Google Scholar]
  • 40.Cheitlin MD, Hutter AM, Brindis RG, et al. The ACC/AHA expert consensus document. Use of Sildenafil (Viagra) in patients with cardiovascular disease. J Am Coll Cardiol 1999;33:273–282. [Erratum in: J Am Coll Cardiol 1999;34:1850] [DOI] [PubMed] [Google Scholar]
  • 41.Ribe N, Rajmil O, Bassas L, et al. Response to intracavernous administration of 3 different drugs in the same group of patients with erectile dysfunction. Arch Esp Urol 2001;54:355–359. [In Spanish] [PubMed] [Google Scholar]
  • 42.Arruda-Olson AM, Mahoney DW, Nehra A, et al. Cardiovascular effects of sildenafil during exercise in men with known or probable coronary artery disease: a randomized crossover trial. JAMA 2002;287:719–725. [DOI] [PubMed] [Google Scholar]
  • 43.Mittleman MA, Glasser DB, Orazem J. Clinical trials of sildenafil citrate (Viagra) demonstrate no increase in risk of myocardial infarction and cardiovascular death compared with placebo. Int J Clin Pract 2003;57:597–600. [PubMed] [Google Scholar]
  • 44.Urciuoli R, Cantisani TA, CarliniI M, et al. Prostaglandin E1 for treatment of erectile dysfunction. In: The Cochrane Library, Issue 3, 2004. Chichester, UK: John Wiley & Sons, Ltd. Search date 2003, primary sources Cochrane Multiple Sclerosis Group Trials Register; the Cochrane Central Register of Controlled Trials; Medline; Embase; reference lists of articles; hand searches of urological and neurological journals, abstracts of conference proceedings, and symposia; and contact with pharmaceutical companies. [Google Scholar]
  • 45.Costabile RA, Spevak M, Fishman IJ, et al. Efficacy and safety of transurethral alprostadil in patients with erectile dysfunction following radical prostatectomy. J Urol 1998;160:1325–1328. [PubMed] [Google Scholar]
  • 46.Shabsigh R, Padma-Nathan H, Gittlemann M, et al. Intracavernous alprostadil alfadex is more efficacious, better tolerated, and preferred over intraurethral alprostadil plus optional ACTIS: a comparative, randomized, crossover, multicenter study. Urology 2000;55:109–113. [DOI] [PubMed] [Google Scholar]
  • 47.Shokeir AA, Alserafi MA, Mutabagani H. Intracavernosal versus intraurethral alprostadil: a prospective randomized study. BJU Int 1999;83:812–815. [DOI] [PubMed] [Google Scholar]
  • 48.Porst H. Transurethral alprostadil with MUSE (medicated urethral system for erection) vs intracavernous alprostadil – a comparative study in 103 patients with erectile dysfunction. Int J Impot Res 1997;9:187–192. [DOI] [PubMed] [Google Scholar]
  • 49.Von Keitz AT, Stroberg P, Bukofzer S, et al. A European multicentre study to evaluate the tolerability of apomorphine sublingual administered in a forced dose-escalation regimen in patients with erectile dysfunction. BJU Int 2002;89:409–415. [DOI] [PubMed] [Google Scholar]
  • 50.Dula E, Keating W, Siami PF, et al. Efficacy and safety of fixed-dose and dose-optimization regimens of sublingual apomorphine versus placebo in men with erectile dysfunction. Urology 2000;56:130–135. [DOI] [PubMed] [Google Scholar]
  • 51.Dula E, Bukofzer S, Perdok R, et al. Double-blind, crossover comparison of 3 mg apomorphine SL with placebo and with 4 mg apomorphine SL in male erectile dysfunction. Eur Urol 2001;39:558–563. [DOI] [PubMed] [Google Scholar]
  • 52.Fagan TC, Buttler S, Marbury T, et al. Cardiovascular safety of sublingual apomorphine in patients on stable doses of oral antihypertensive agents and nitrates. Am J Cardiol 2001;88:760–766. [DOI] [PubMed] [Google Scholar]
  • 53.Stief C, Padley RJ, Perdok RJ, et al. Cross-study review of the clinical efficacy of apomorphine SL 2 and 3 mg: pooled data from three placebo-controlled, fixed-dose crossover studies. Eur Urol 2002;1:12–20. [Google Scholar]
  • 54.Heaton JP. Apomorphine: an update of clinical trial results. Int J Impot Res 2000;12(Suppl):S67–S73. [DOI] [PubMed] [Google Scholar]
  • 55.Jiang H, Xu QQ, Hong K, et al. Efficacy and safety of PGE1 cream in the treatment of erectile dysfunction. Zhonghua Nan Ke Xue 2003;9:97–99. [In Chinese] [PubMed] [Google Scholar]
  • 56.Goldstein I, Payton TR, Schechter PJ. A double-blind, placebo-controlled, efficacy and safety study of topical gel formulation of 1% alprostadil (Topiglan) for the in-office treatment of erectile dysfunction. Urology 2001;57:301–305. [DOI] [PubMed] [Google Scholar]
  • 57.Padma-Nathan H, Steidle C, Salem S, et al. The efficacy and safety of a topical alprostadil cream, Alprox-TD, for the treatment of erectile dysfunction: two phase 2 studies in mild-to-moderate and severe ED. Int J Impot Res 2003;15:10–17. [DOI] [PubMed] [Google Scholar]
  • 58.Keogh EJ, Watters GR, Earle CM, et al. Treatment of impotence by intrapenile injections. A comparison of papaverine versus papaverine and phentolamine: a double-blind, crossover trial. J Urol 1989;142:726–728. [DOI] [PubMed] [Google Scholar]
  • 59.Pastorini S, Marino G, Cocimano V, et al. Complications of intracavernous pharmacologic infusion in impotence. Long-term results. Minerva Urol Nefrol 1993;45:109–112. [In Italian] [PubMed] [Google Scholar]
  • 60.Gasser TC, Roach RM, Larsen EH, et al. Intracavernous self-injection with phentolamine and papaverine for the treatment of impotence. J Urol 1987;137:678–680. [DOI] [PubMed] [Google Scholar]
  • 61.Levine SB, Althof SE, Turner LA, et al. Side effects of self-administration of intracavernous papaverine and phentolamine for the treatment of impotence. J Urol 1989;141:54–57. [DOI] [PubMed] [Google Scholar]
  • 62.Montague DK, Angermeier KW. Contemporary aspects of penile prosthesis implantation. Urol Int 2003;70:141–146. [DOI] [PubMed] [Google Scholar]
  • 63.Linet OI, Ogrinc FG. Efficacy and safety of intracavernosal alprostadil in men with erectile dysfunction. N Engl J Med 1996;334:873–877. [DOI] [PubMed] [Google Scholar]
  • 64.Bechara A, Casabe A, Cheliz G, et al. Comparative study of papaverine plus phentolamine versus prostaglandin E1 in erectile dysfunction. J Urol 1997;157:2132–2134. [PubMed] [Google Scholar]
  • 65.Earle CM, Keogh EJ, Wisniewski ZS, et al. Prostaglandin E1 therapy for impotence, comparison with papaverine. J Urol 1990;143:57–59. [DOI] [PubMed] [Google Scholar]
  • 66.Kattan S, Collins JP, Mohr D. Double-blind, cross-over study comparing prostaglandin E1 and papaverine in patients with vasculogenic impotence. Urology 1991;37:516–518. [DOI] [PubMed] [Google Scholar]
  • 67.Mahmoud KZ, el Dakhli MR, Fahmi IM, et al. Comparative value of prostaglandin E1 and papaverine in treatment of erectile failure: double-blind crossover study among Egyptian patients. J Urol 1992;147:623–626. [DOI] [PubMed] [Google Scholar]
  • 68.Bechara A, Casabe A, Cheliz G, et al. Prostaglandin E1 versus mixture of prostaglandin E1, papaverine and phentolamine in nonresponders to high papaverine plus phentolamine doses. J Urol 1996;155:913–914. [PubMed] [Google Scholar]
  • 69.Soderdahl DW, Thrasher JB, Hansberry KL. Intracavernosal drug-induced erection therapy versus external vacuum devices in the treatment of erectile dysfunction. Br J Urol 1997;79:952–957. [DOI] [PubMed] [Google Scholar]
  • 70.Chiang HS, Wu CC, Wen TC. 10 years of experience with penile prosthesis implantation in Taiwanese patients. J Urol 2000;163:476–480. [PubMed] [Google Scholar]
  • 71.Wylie KR, Jones RH, Walters S. The potential benefit of vacuum devices augmenting psychosexual therapy for erectile dysfunction: a randomized controlled trial. J Sex Marital Ther 2003;29:227–236. [DOI] [PubMed] [Google Scholar]
  • 72.Turner LA, Althof SE, Levine SB, et al. Treating erectile dysfunction with external vacuum devices: impact upon sexual, psychological and marital functioning. J Urol 1990;144:79–82. [DOI] [PubMed] [Google Scholar]
  • 73.Price SC, Reynolds BS, Cohen BD, et al. Group treatment of erectile dysfunction for men without partners: a controlled evaluation. Arch Sex Behav 1981;10:253–268. [DOI] [PubMed] [Google Scholar]
  • 74.Stravynski A, Gaudette G, Lesage A, et al. The treatment of sexually dysfunctional men without partners: a controlled study of three behavioural group approaches. Br J Psychiatry1997;170:338–344. [DOI] [PubMed] [Google Scholar]
  • 75.Hong B, Ji YH, Hong JH, et al. A double-blind crossover study evaluating the efficacy of Korean red ginseng in patients with erectile dysfunction: a preliminary report. J Urol 2002;168:2070–2073. [DOI] [PubMed] [Google Scholar]
  • 76.Ernst E, Pittler MH. Yohimbine for erectile dysfunction: a systematic review and meta-analysis of randomized clinical trials. J Urol 1998;159:433–436. Search date 1997; primary sources Medline, Embase, The Cochrane Library, and hand searches of references. [DOI] [PubMed] [Google Scholar]
  • 77.Kunelius P, Hakkinen J, Lukkarinen O. Is high-dose yohimbine hydrochloride effective in the treatment of mixed-type impotence? A prospective, randomized, controlled double-blind crossover study. Urology 1997;49:441–444. [DOI] [PubMed] [Google Scholar]
  • 78.Seftel AD, Wilson SK, Knapp PM, et al. The efficacy and safety of tadalafil in United States and Puerto Rican men with erectile dysfunction.[erratum appears in J Urol. 2005 Feb;173(2):664]. Journal of Urology 2004/8;172:652-657 [DOI] [PubMed] [Google Scholar]
  • 79.DeBusk RF, Pepine CJ, Glasser DB, et al. Efficacy and safety of sildenafil citrate in men with erectile dysfunction and stable coronary artery disease.[erratum appears in Am J Cardiol. 2004 Aug 15;94(4):543-4]. American Journal of Cardiology 2004/1/15;93:147-153 [DOI] [PubMed] [Google Scholar]
BMJ Clin Evid. 2006 Mar 1;2006:1803.

Sildenafil

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo: Sildenafil is more effective at increasing the proportion of men who experience successful intercourse and improved erections ( moderate-quality evidence ). Compared with intracavernosal alprostadil: Sildenafil and intracavernosal alprostadil are equally effective at 4–9 months (average 6 months) at improving erections sufficient for successful intercourse (moderate-quality evidence). ADVERSE EFFECTS Sildenafil has been associated with headache, flushing, visual disturbances and dyspepsia. Sildenafil is contraindicated in men using oral nitrates concomitantly, in whom deaths have been reported.

Benefits

We found one systematic review (search date 2000, 27 RCTs) and 15 subsequent RCTs comparing sildenafil versus placebo, and one RCT comparing sildenafil versus intracavernosal alprostadil.

Sildenafil versus placebo in men with erectile dysfunction of any cause:

The systematic review found that, compared with placebo, flexible “as needed” dosing of sildenafil significantly increased the proportion of men who experienced at least one episode of successful intercourse (14 RCTs, 2283 men with any cause of erectile dysfunction; at least 1 episode of successful intercourse in 4 weeks preceding end of treatment assessment: 83% with sildenafil v 45% with placebo; RR 1.8, 95% CI 1.7 to 1.9). In trials that evaluated fixed doses of sildenafil (7 RCTs), efficacy in treatment responders was slightly higher on higher doses (50–100 mg) compared with doses less than 25 mg (successful sexual intercourse, mean percentage of attempts/person: 43% with sildenafil 25 mg v 17% with placebo; WMD 26, 95% CI 18 to 35; 50% with sildenafil 50 mg v 14% with placebo; WMD 36, 95% CI 30 to 42; 51% with sildenafil 100 mg v 14% with placebo; WMD 36, 95% CI 31 to 42). Fifteen subsequent RCTs all found that sildenafil significantly improved sexual function compared with placebo. The largest of these RCTs found that, compared with placebo, sildenafil as needed significantly increased the proportion of people who had successful intercourse within 20 minutes over 4 weeks (228 men, mean age 60 years, mean duration of erectile dysfunction 7 years; 51% with sildenafil 100 mg as needed v 30% with placebo; P < 0.05). The median time to onset of erections that resulted in successful sexual intercourse was 36.3 minutes with sildenafil 100 mg compared with 140.7 minutes with placebo (RR 3.49, 95% CI 2.43 to 5.01).

Sildenafil versus intracavernosal alprostadil in men with erectile dysfunction of any cause:

We found one small RCT (54 men) that compared sildenafil versus intracavernosal alprostadil injections. Sildenafil and alprostadil injections both improved erections sufficiently for successful intercourse over 4–9 months (average 6 months) of treatment, with no significant difference between treatments (AR: 80% with sildenafil v 83.3% with alprostadil injections; P > 0.05).

Harms

Sildenafil versus placebo in men with erectile dysfunction of any cause:

The systematic review found that in a subset of 14 flexible dose trials (3780 men), sildenafil significantly increased the risk of at least one adverse effect compared with placebo (3780 men with any cause of erectile dysfunction; AR: 48% with sildenafil v 36% with placebo; RR 1.4, 95% CI 1.3 to 1.6). Adverse effects included headache (11% with sildenafil v 4% with placebo; RR 2.6, 95% CI 1.8 to 3.7), flushing (12% with sildenafil v 2% with placebo; RR 5.8, 95% CI 3.4 to 10), dyspepsia (5% with sildenafil v 1% with placebo; RR 3.8, 95% CI 2.2 to 6.6), and visual disturbance (3.0% with sildenafil v 0.8% with placebo; RR 3.1, 95% CI 1.8 to 5.4). Similar proportions of those allocated to sildenafil or placebo discontinued treatment because of adverse effects (1.3% with sildenafil v 1.2% with placebo; RR 1.3, 95% CI 0.7 to 2.3). The data from fixed dose trials in this systematic review indicate that all these adverse effects were more frequent at higher doses of sildenafil and were mild to moderate in severity. One subsequent RCT (236 men with any cause of erectile dysfunction) found that sildenafil was associated with facial flushing (25.2%), dizziness (6.7%), headache (5.9%), and palpitations (3.4%). A second RCT found that headache (20% with sildenafil v 6% with placebo), flushing (15% with sildenafil v 1% with placebo), dyspepsia (15% with sildenafil v 0% with placebo), and abnormal perception of colour or brightness (8% with sildenafil v 1% with placebo; significance assessment not performed for any adverse effect) were more common with sildenafil than placebo. Four other RCTs reported on harms and found similar results.

Comment

Many of the trials in the systematic review and all of the subsequent RCTs did not describe the procedure used to generate the randomisation sequence or conceal allocation of treatment assignment. This may reflect inadequate reporting or inadequate randomisation. Trials with improper randomisation, especially those with poor concealment of allocation to treatment assignments, have been shown to overestimate treatment effects.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Tadalafil

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men with erectile dysfunction of any cause: Tadalafil may be more effective at improving erections and successful sexual intercourse which may persist for up to 36 hours ( very low-quality evidence ). ADVERSE EFFECTS Tadalafil has been associated with headache, muscle pain, back ache, dyspepsia, and flushing. Tadalafil is contraindicated in people receiving nitrates concomitantly because of the risk of potentially life threatening hypotension.

Benefits

We found no systematic review.

Tadalafil versus placebo in men with erectile dysfunction of any cause:

We found one systematic review of 11 RCTs (2102 men aged 22–88 years [mean 56 years], mild to severe erectile dysfunction for at least 1 year resulting from organic, psychogenic, or mixed causes; given tadalafil 10 mg [312 men], tadalafil 20 mg [1143 men], or placebo [638 men] over 12 weeks, sponsored by the manufacturers and conducted worldwide at 174 centres from April 1999 to February 2003; see comment below). Mean improvement in erectile function, assessed using International Index of Erectile Function Erectile Function domain (IIEF EF domain) was significantly greater for both doses of tadalafil compared with placebo (2036 men, mean change in IIEF EF domain score: 6.5 with tadalafil 10 mg v 8.6 with tadalafil 20 mg v 0.9 with placebo; P < 0.001 for both doses of tadalafil v placebo). Both doses of tadalafil significantly increased the proportion of sexual attempts leading to intercourse completion, as assessed by the Sexual Encounter Profile (SEP) question 3 (SEP-Q3; 2055 men, mean change from baseline [range: 22–24%] for intercourse completion [SEP-Q3]: 34% with tadalafil 10 mg v 46% with tadalafil 20 mg v 8% with placebo; P < 0.001 for both doses v placebo). Both doses of tadalafil also significantly improved erections compared with placebo, as assessed by the Global Assessment Questionnaire (2055 men, proportion reporting improved erections: 71% with tadalafil 10 mg v 84% with tadalafil 20 mg v 33% with placebo; P < 0.001 for both doses of tadalafil v placebo). We found three subsequent RCTs that compared tadalafil with placebo. One RCT (179 men aged 21–72 years [mean 56 years], with any cause of erectile dysfunction) compared tadalafil in fixed doses of 2, 5, 10, or 25 mg versus placebo over a 3 week period. All doses of tadalafil significantly improved International Index of Erectile Function (IIEF) question 3 scores compared with placebo (179 men, mean change from baseline score [range: 2.36–3.1] for IIEF question 3: +0.6 with 2 mg tadalafil v +1.2 with 5 mg tadalafil v +1.0 with 10 mg tadalafil v +1.3 with 25 mg tadalafil v –0.3 with placebo; P < 0.005 for tadalafil 2 mg v placebo; P < 0.0005 for 5, 10, and 25 mg tadalafil v placebo). All doses of tadalafil, except 2 mg, significantly improved IIEF question 4 scores compared with placebo (179 men, mean change from baseline score [range: 1.9–2.5%] for IIEF question 4: +0.8 with 2 mg tadalafil v +1.4 with 5 mg tadalafil v +1.7 with 10 mg tadalafil v +1.7 with 25 mg tadalafil v +0.2 with placebo; P < 0.0005 for 5, 10, and 25 mg tadalafil v placebo). Tadalafil was associated with higher levels of overall satisfaction with the sexual experience for both men and their partners compared with placebo (179 couples, proportion of people reporting overall satisfaction; men: 58% with tadalafil v 16.6% with placebo; P ≤ 0.05; partners: 63% with tadalafil v 19.6% with placebo; significance assessment not performed). The second subsequent RCT (207 men, with mostly organic causes for erectile dysfunction) compared tadalafil 20 mg taken as needed over 12 weeks (159 men) with placebo (48 men). Tadalafil significantly improved successful erection and sexual intercourse scores compared with placebo, as assessed by the SEP-Q2 and SEP-Q3 (mean change from SEP-Q2 penetration baseline score [range: 40.5–45.2%]: +31.6 with tadalafil v +2.3 with placebo; P < 0.0001; mean change from SEP-Q3 intercourse baseline score [range: 19.1–20.7%]: 43.6 with tadalafil v 3.5 with placebo; P < 0.0001). Compared with placebo, tadalafil also significantly increased both the number of intercourse attempts initiated between 4–36 hours after dosing and the proportion occurring between 4–36 hours that were successful (proportion of attempts at intercourse within 4–36 hours after dosing: 2446/4433 [55.2%] with tadalafil v 350/1005 [34.8%] with placebo; P = 0.001; proportion of successful attempts at intercourse within 4–36 hours after dosing: 1728/2988 [57.8%] with tadalafil v 63/253 [26.8%] with placebo; P < 0.001). The third subsequent RCT (348 men, mean age 57 years, with various causes of erectile dysfunction) compared tadalafil 20 mg with placebo taken 24 or 36 hours before sexual intercourse. Tadalafil significantly increased the proportion of successful attempts compared with placebo at both 24 and 36 hours (348 men, proportion of successful intercourse attempts at 24 hours: 120/227 [52.9%] with tadalafil v 72/247 [29.1%] with placebo; at 36 hours: 132/223 [59.2%] with tadalafil v 60/212 [28.3%] with placebo; P < 0.001 for both time periods v placebo).

Tadalafil versus sildenafil in men with erectile dysfunction of any cause:

We found one RCT (215 men aged 18–65 years, minimum 3 months of erectile dysfunction, no prior use of tadalafil, 15.3% had received an inadequate trial of 50 mg sildenafil) that evaluated patient preference for either tadalafil or sildenafil in a double blind crossover trial. The RCT compared tadalafil 20 mg versus sildenafil 50 mg as needed for 4 weeks, and after 1–2 weeks of drug washout crossed them over to the alternative treatment for 4 weeks. Of 190/215 (88%) men who expressed a preference, a significantly higher proportion preferred tadalafil (AR: 126/190 [66.3%] for tadalafil 20 mg v 64/190 [33.7%] for sildenafil 50 mg; P < 0.001).

Harms

Tadalafil versus placebo in men with erectile dysfunction of any cause:

The integrated results of 11 RCTs suggested that a higher proportion of men treated with tadalafil experienced one or more adverse effects compared with placebo (AR: 39% with placebo v 58% with tadalafil 10 mg and 51% with tadalafil 20 mg; significance assessment not performed). Although absolute numbers were few, significantly more men withdrew because of adverse effects with tadalafil than with placebo (AR: 5/321 [1.6%] with tadalafil 10 mg v 36/1143 [3.2%] with tadalafil 20 mg v 8/638 [1.3%] with placebo; P = 0.026). Adverse effects included headache, dyspepsia, back pain, nasopharyngitis, myalgia, flushing, nasal congestion, and limb pain. One subsequent RCT found that frequency of adverse events with tadalafil increased with dose (adverse events: 17.1% with tadalafil 2 mg v 36.1% with tadalafil 25 mg; significance not assessed). This and two other subsequent RCTs found that more men taking tadalafil withdrew because of adverse effects compared with placebo. Chest pain requiring admission to hospital occurred in 2/207 [1%] men taking tadalafil 20 mg.

Tadalafil versus sildenafil in men with erectile dysfunction of any cause:

In the RCT, the most frequently reported adverse event was headache (24/215 [11.2%] with tadalafil v 19/215 [8.8%] with sildenafil; significance assessment not performed). Other adverse events included dyspepsia, nasopharyngitis, flushing, myalgia, and nasal congestion.

Comment

The included trials in the systematic review were clinically heterogeneous in aetiology of erectile dysfunction and age, and lacked homogenous study design or outcome assessments. Statistical heterogeneity was not formally assessed. None of the RCTs described the procedure used to generate the randomisation sequence or to conceal allocation. However, the trials comparing patient preference for tadalafil or sildenafil were limited by the inclusion of people with prior experience with sildenafil, the lack of provision to use lower doses of tadalafil than 20 mg, and the limitation of upward dose titration of sildenafil to only 35% of those requesting it. We found no reports of visual adverse events with tadalafil.

Clinical guide:

Tadalafil provides greater flexibility in timing of sexual intercourse and requires no dietary restrictions compared with sildenafil that has a 4 hour window after dose and decreased efficacy with grapefruit juice or after a high fat meal.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Vardenafil

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men with erectile dysfunction of any cause: Vardenafil may be more effective at improving erections and at increasing rates of successful intercourse ( low-quality evidence ). ADVERSE EFFECTS Vardenafil has been associated with headache, flushing, and dyspepsia. Vardenafil is contraindicated in people receiving nitrates concomitantly becuase of the risk of potentially life threatening hypotension.

Benefits

Vardenafil versus placebo in men with erectile dysfunction of any cause:

We found one systematic review (search date 2002, 5 RCTs, 1900 men) and one subsequent RCT. The systematic review found two RCTs with clinical end points. The first RCT identified by the review found that, compared with placebo, vardenafil 5, 10, and 20 mg, taken 1 hour before attempts at sexual intercourse significantly improved erections and successful sexual intercourse as measured by questions 3 and 4 on the International Index of Erectile Function (IIEF) over 12 weeks (601 men, mean age 52.2 years with erectile dysfunction for ≥ 6 months; mean baseline erectile function domain score 14.0, mean improvement in erectile function domain score at 12 weeks: 6.7 with vardenafil 5 mg v 8.0 with vardenafil 10 mg v 9.0 with vardenafil 20 mg v 1.6 with placebo; P < 0.001 for all comparisons v placebo). The second RCT identified by the review found that, compared with placebo, vardenafil 5, 10, and 20 mg significantly improved scores on the erectile function domain of the IIEF at 12 weeks but only vardenafil 10 and 20 mg improved scores at 26 weeks (805 men, mean age 57.2 years; mean baseline erectile function domain score 13.1, mean improvement in erectile function domain scores at 12 weeks: 5.9 with vardenafil 5 mg v 7.2 with vardenafil 10 mg v 8.6 with vardenafil 20 mg v 1.4 with placebo; P < 0.001 for all comparisons v placebo; erectile function domain scores at 26 weeks: P < 0.001 for 10 and 20 mg v placebo). Vardenafil 5, 10, and 20 mg also significantly improved successful erections and sexual intercourse as assessed by questions 2 and 3 on the Sexual Encounter Profile diaries (P < 0.0001 for each comparison v placebo) as well as the proportion of men responding yes to the Global Efficacy Question (P < 0.001 for 5, 10, and 20 mg of vardenafil v placebo). We found one subsequent RCT. It found that flexible doses of vardenafil significantly improved scores on the erectile function domain of the IIEF compared with placebo (323 men, mean age 54 years, 50–54% with prior sildenafil use, mean baseline erectile function domain score 13, 5–20 mg as desired for 8 weeks after initial 4 weeks of 10 mg; mean improvement in erectile function domain score at 12 weeks: 3.2 with vardenafil v 1.9 with placebo; P < 0.005). Vardenafil also improved erections as assessed by questions 2 (penetration) and 3 (successful intercourse) of the Sexual Encounter Profile diaries (mean per person success with penetration at baseline: 38–39% with vardenafil v 44–48% with placebo; mean per person success with successful intercourse at baseline: 14% with vardenafil v 18–20% with placebo; mean per person success with penetration at week 4: 73% with vardenafil v 41% with placebo; week 8: 84% with vardenafil v 49% with placebo; week 12: 80% with vardenafil v 46% with placebo; P < 0.001 for all comparisons with placebo; mean per person success with intercourse at week 4: 58% with vardenafil v 22% with placebo; week 8: 71% with vardenafil v 31% with placebo; week 10: 74% with vardenafil v 34% with placebo; P < 0.001 for all comparisons with placebo).

Harms

The systematic review found that adverse effects with vardenafil 5–40 mg were reported in 22–61% of people in four RCTs. In the largest RCT (762 men in safety analysis), headache was the most common (10–21%) followed by rhinitis (9–17%), flushing (5–13%), and dyspepsia (1–6%). The rates of adverse events with placebo were not reported in the systematic review. The review also included a crossover, single dose study (41 men with stable angina) to assess exercise tolerance in men on vardenafil 10 mg or placebo. Vardenafil did not reduce symptom limited exercise time or time to awareness of angina compared with placebo (P = 0.39–0.59). Vardenafil use did increase time to ST segment depression compared with placebo (381 seconds with vardenafil v 334 seconds with placebo; P = 0.0004). The review found that in one drug interaction trial in men with hypertension, the addition of vardenafil 20 mg to the vasodilator nifedipine did not result in clinically important changes in blood pressure (data not reported). A subsequent RCT found that, compared with placebo, vardenafil significantly increased the proportion of men who experienced adverse effects (adverse effects reported: 14% with vardenafil 5 mg v 22% with vardenafil 10 mg v 11% with vardenafil 20 mg v 5% with placebo; significance assessment not performed). Overall 45/157 (29%) men on vardenafil experienced at least one adverse effect compared with 5% on placebo. Drug related adverse effects reported by more than 2% of men included flushing (18/157 [11%] with vardenafil v 0/164 [0%] with placebo), headache (15/157 [10%] with vardenafil v 3/164 [2%] with placebo), rhinitis (8/157 [5%] with vardenafil v 1/164 [1%] with placebo), dyspepsia (4/157 [3%] with vardenafil v 0/164 [0%] with placebo), and dizziness (3/157 [2%] with vardenafil v 1/164 [1%] with placebo). The same contraindications for concomitant nitrate use apply to vardenafil as with other phosphodiesterase 5 inhibitors (see harms of sildenafil versus placebo in men with heart disease).

Comment

The systematic review restricted electronic searches to Medline and to articles in English. Trials were not assessed for quality or if they were, this was not described. All the trials excluded men who had failed to respond to sildenafil though 49–71% reported previous sildenafil use and were not treatment naive. Using previous treatment responders is likely to bias the results in favour of the drug. Data were not synthesised in a meta-analysis but reported in a narrative manner. Vardenafil is reportedly the most selective and potent phosphodiesterase 5 inhibitor, but we did not find any comparisons of vardenafil with sildenafil, tadalafil, or other treatments for erectile dysfunction. We found no reports of visual adverse events with vardenafil.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Sildenafil

Summary

IMPROVEMENT IN SEXUAL DYSFUNCTION Compared with placebo in men with diabetes: Sildenafil is more effective at improving erections and at increasing successful intercourse in diabetic men ( high-quality evidence ). ADVERSE EFFECTS Sildenafil has been associated with headache, flushing, visual disturbances and dyspepsia. Sildenafil is contraindicated in men using oral nitrates concomitantly, in whom deaths have been reported.

Benefits

Sildenafil versus placebo in men with diabetes:

We found one systematic review (search date 2000, 27 RCTs). The systematic review identified two RCTs restricted to men with diabetes and 14 RCTs that provided subgroup analysis in men with diabetes. Based on subgroup analysis, the review found that, compared with placebo, sildenafil significantly increased successful erections and successful intercourse (1019 men with diabetes; AR for erections: 63% with sildenafil v 19% with placebo; RR 3.0, 95% CI 2.5 to 3.7; 551 men with diabetes; AR for mean percentage of successful intercourse attempts: 44% with sildenafil v 16% with placebo; WMD 27, 95% CI 20 to 34). We found four subsequent RCTs. The first subsequent RCT found that, compared with placebo, sildenafil 25–100 mg significantly improved participant rated erections and the capacity to obtain and maintain an erection as measured by questions 3 and 4 of the International Index of Erectile Dysfunction after 12 weeks (219 men with erectile dysfunction and diabetes; AR for improved participant rated erections: 64.6% with sildenafil v 10.5% with placebo; CI presented graphically; P less than 0.0001; mean improvement in question 3 score after 12 weeks: 3.42 with sildenafil v 1.86 with placebo; mean improvement in question 4 score after 12 weeks: 3.35 with sildenafil v 1.84 with placebo; P less than 0.0001 for both comparisons). The second subsequent RCT also found that, compared with placebo, sildenafil significantly improved scores on questions 3 and 4 of the International Index of Erectile Function (IIEF) compared after 12 weeks (188 men with erectile dysfunction and diabetes; mean question 3 score after 12 weeks: 3.61 with sildenafil v 2.71 with placebo; P = 0.001; mean question 4 score after 12 weeks: 3.25 with sildenafil v 2.19 with placebo; P = 0.001). Sildenafil also increased the proportion of successful attempts at intercourse compared with placebo, although the result was of borderline significance (results presented graphically; P = 0.051). The third subsequent RCT found that, compared with placebo, sildenafil significantly improved the scores on questions 3 and 4 of the IIEF (112 men with erectile dysfunction and diabetes: mean question 3 score: 3.0 with sildenafil v 1.7 with placebo; P < 0.0001; mean question 4 score: 2.6 with sildenafil v 1.6 with placebo; P less than 0.0001).The fourth subsequent RCT found that, compared with placebo, sildenafil significantly increased the proportion of people who reported at least one successful attempt at intercourse (282 men with erectile dysfunction and diabetes, 262 completers; at least 1 successful attempt at intercourse: 59% with sildenafil v 21% with placebo; P less than 0.002).

Harms

Sildenafil versus placebo in men with diabetes:

One RCT (188 men with diabetes) found that, compared with placebo, sildenafil increased adverse events such as headache (20% with sildenafil v 8% with placebo), flushing (18% with sildenafil v 3% with placebo), and dyspepsia (32% with sildenafil v 8% with placebo; significance assessment not performed). Another small RCT (60 men) reported similar results (headache: 42% with sildenafil v 15% with placebo; P less than 0.001; dyspepsia: 32% with sildenafil v 8% with placebo; P less than 0.001). It found no significant difference between sildenafil and placebo for other adverse effects, including myalgia, nasal congestion, visual disturbance, and dizziness. One placebo controlled RCT found that 4/137 (2.7%) men on sildenafil developed chest pain and 2/137 (1.5%) developed myocardial infarction.

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Tadalafil

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men with diabetes: Tadalafil is more effective at improving erections and sexual functioning in diabetic men ( high-quality evidence ). ADVERSE EFFECTS Tadalafil has been associated with headache, muscle pain, back ache, dyspepsia, and flushing. Tadalafil is contraindicated in people receiving nitrates because of the risk of potentially life threatening hypotension.

Benefits

Tadalafil versus placebo in men with diabetes:

We found one RCT (216 men with Type 1 or Type 2 diabetes and a minimum 3 month history of erectile dysfunction) that compared tadalafil 10 mg versus 20 mg versus placebo taken up to once daily without restrictions on food or alcohol intake for 12 weeks. Both doses of tadalafil significantly improved erections and sexual functioning, as assessed by IIEF EF (mean change in IIEF EF domain score: 6.4 with tadalafil 10 mg v 7.3 with tadalafil 20 mg v 0.1 with placebo; P less than 0.001 for both doses v placebo). For men taking concomitant antihypertensive medication, tadalafil 20 mg was associated with a better response than tadalafil 10 mg or placebo, assessed by IIEF EF (mean change in IIEF EF domain score: –1.8 with placebo v +3.9 with tadalafil 10 mg v +9.5 with tadalafil 20 mg; P value for tadalafil 10 mg v placebo significance assessment not performed, P < 0.001 for tadalafil 20 mg v placebo), Sexual Encounter Profile question 2 (successful penetration) and question 3 (successful intercourse) (change from SEP-Q2 penetration baseline score [range not reported]: +22.2% with tadalafil 10 mg v +30.6% with tadalafil 20 mg v –4.1% with placebo; P less than 0.001 for both doses v placebo; change from SEP-Q3 intercourse baseline score [range not reported]: 28.4% with tadalafil 10 mg v 29.1% with tadalafil 20 mg v 1.9 with placebo; P less than 0.001 for both doses v placebo).

Harms

Tadalafil versus placebo in men with diabetes:

The RCT found a generally higher incidence of adverse effects with tadalafil compared with placebo but this difference was only significant for dyspepsia (8/73 [11.0%] with tadalafil 10 mg v 8/72 [11.1%] with tadalafil 20 mg v 0/71 [0%] with placebo; P = 0.005).

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Vardenafil

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men with diabetes: Vardenafil is more effective at improving erections and sexual functioning in diabetic men at 12 weeks ( high-quality evidence ). ADVERSE EFFECTS Vardenafil has been associated with headache, flushing, and dyspepsia. Vardenafil is contraindicated in people receiving nitrates because of the risk of potentially life threatening hypotension.

Benefits

Vardenafil versus placebo in men with diabetes:

We found one systematic review (search date 2000, 27 RCTs). The systematic review identified one RCT comparing vardenafil versus placebo in men with diabetes (452 men, glycosylated haemoglobin < 12%, erectile dysfunction of > 6 months) that compared vardenafil 10 or 20 mg with placebo over 12 weeks followed by an extension phase of 3 months (343 men, 118 continued on vardenafil 10 mg, 116 continued on 20 mg and of 106 men originally on placebo, 55 took vardenafil 10 mg and 51 took vardenafil 20 mg). It found that vardenafil significantly improved scores on the erectile function domain of the IIEF after 12 weeks compared with placebo (mean scores for erectile function at baseline 11.3; mean scores for erectile function at 12 weeks: 13 with placebo v 17 with vardenafil 10 mg v 19 with vardenafil 20 mg; P less than 0.0001 for both comparisons versus placebo). At the end of 3 months of extension treatment, scores on the erectile function domain improved for men changed from placebo to vardenafil and remained unchanged for those continuing on vardenafil (figures not reported; significance assessment not performed).

Harms

The systematic review found that adverse effects with vardenafil 5–40 mg were reported in 22–61% of people in four RCTs. In the largest RCT (762 men in safety analysis), headache was the most common (10–21%) followed by rhinitis (9–17%), flushing (5–13%), and dyspepsia (1–6%). The rates of adverse events with placebo were not reported in the systematic review.

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Phosphodiesterase inhibitors

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men with heart disease: Sildenafil is more effective at improving the proportion of men with successful erections and intercourse at 12 weeks ( high-quality evidence ). ADVERSE EFFECTS Compared with placebo: In men with ischaemic heart disease and who are not receiving nitrates, sildenafil increases the proportion of men with angina (high-quality evidence). NOTE Sildenafil is contraindicated in men using oral nitrates concomitantly, in whom deaths have been reported.

Benefits

Sildenafil versus placebo in men with heart disease:

We found one systematic review (search date 2000, 27 RCTs). The systematic review included a subgroup analysis of people with ischaemic heart disease. It found that sildenafil significantly improved the proportion of men with successful erections and with successful sexual intercourse compared with placebo (373 men with heart disease and erectile dysfunction; AR for successful erections: 63% with sildenafil v 20% with placebo; relative benefit increase 2.6, 95% CI1.8 to 3.8; 202 men, AR for mean percentage of successful sexual intercourse attempts: 42% with sildenafil v 14% with placebo, WMD 23.8, 95% CI 2.1 to 45.6). We found two subsequent RCTs that compared sildenafil versus placebo in men with heart disease. The first subsequent RCT found that compared with placebo, flexible doses of sildenafil significantly improved mean scores for question 3 and 4 of the IIEF after 12 weeks (224 men aged > 40 years with heart disease on a variety of antihypertensive drugs barring nitrates and erectile dysfunction; mean question 3 score after 12 weeks: 3.7 with sildenafil v 2.2 with placebo; mean question 4 score after 12 weeks: 3.3 with sildenafil v 1.9 with placebo; P = 0.0001 for both comparisons). Sildenafil significantly increased erectile function as assessed by the Global Efficacy Question compared with placebo (71% with sildenafil and 24% with placebo; P = 0.0001). The second subsequent RCT found that flexible doses of sildenafil significantly improved scores on questions 3 and 4 of the IIEF compared with placebo after 12 weeks (142 men aged 39–82 years with any cause of erectile dysfunction [mean duration 5 years] and stable coronary heart disease [mean duration 7 years] including 50% with previous myocardial infarctions [> 8 weeks] and 50% with previous coronary angioplasties, coronary artery bypass grafting, or both; mean improvement in question 3 score after 12 weeks: 3.5 with sildenafil v 2.7 with placebo; mean improvement in question 4 score after 12 weeks: 3.3 with sildenafil v 2.3 with placebo; P less than 0.006 for both comparisons).

Harms

Sildenafil versus placebo in men with heart disease:

An important contraindication to prescribing sildenafil is concomitant use of oral nitrates within 24 hours as the combination could potentially result in precipitous hypotension. Two RCTs (264 men) that assessed haemodynamic performance in people with heart disease on a variety of antihypertensive drugs other than nitrates found no clinically significant changes in blood pressure with sildenafil. The first RCT (224 men) evaluated the effects of sildenafil in men aged 40 years or over with cardiovascular disease on a variety of antihypertensive drugs barring nitrates. Apart from flushing (17% with sildenafil v 2% with placebo), no other cardiovascular adverse events were reported. The second small RCT (105 men) evaluated the cardiovascular effects of sildenafil during exercise in men with coronary heart disease. It found no effect on symptoms, presence, and extent of ischaemia induced by exercise. The systematic review (27 RCTs, 4240 men on sildenafil and 2707 men on placebo) found no significant difference between sildenafil and placebo in mortality (4/4240 [0.1%] with sildenafil v 2/2707 [0.1%] with placebo) or serious cardiovascular morbidity (myocardial infarction: 6/4240 [0.1%] with sildenafil v 6/2707 [0.2%] with placebo). A subgroup analysis restricted to those with ischaemic heart disease not taking nitrates found that, compared with placebo, sildenafil significantly increased the proportion of men with angina significantly more than placebo (24 RCTs, 664 men with ischaemic heart disease not taking nitrates; AR: 2.4% with sildenafil v 0.4% with placebo; P = 0.06). There were no deaths attributed to sildenafil treatment in the two RCTs that evaluated the cardiovascular effects of sildenafil in men with heart disease. One study pooled data regarding myocardial infarction and cardiovascular causes of death from more than 120 clinical trials (RCTs, open label parent studies, and open label extension studies) of sildenafil citrate conducted from 1993 to 2001. The use of sildenafil was not associated with an increase in the risk of myocardial infarction or cardiovascular causes of death (0.91/100 person years of follow up, 95% CI 0.52 to 1.48 with sildenafil v 0.84/100 person years of follow up, 95% CI 0.39 to 1.60 with placebo; RR 1.08, 95% CI 0.45 to 2.77; P = 0.88).

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Phosphodiesterase inhibitors

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men with spinal cord injury: Sildenafil is more effective at improving erections in men with spinal cord injury ( high-quality evidence ). ADVERSE EFFECTS Sildenafil has been associated with dizziness and blood pressure pressure changes in men with spinal cord injuries.

Benefits

Sildenafil versus placebo in men with spinal cord injury:

We found one systematic review (search date 2000, 27 RCTs). The systematic review identified one RCT restricted to men with spinal cord injury. It found that sildenafil significantly improved erections compared with placebo (205 men with erectile dysfunction and spinal cord injury; AR: 83% with sildenafil v 12% with placebo; RR 7.2, 95% CI 4.7 to 10.9).

Harms

Sildenafil versus placebo in men with spinal cord injury:

One placebo controlled, crossover RCT (23 men) evaluated the cardiovascular response to sildenafil in men with complete spinal cord injury at or above the sixth thoracic level. It found that sildenafil significantly decreased systolic blood pressure in men with cervical spine injury compared with placebo (12 men with cervical spine injury and erectile dysfunction; mean change in supine systolic blood pressure at 1 hour: 13.8 mm Hg with sildenafil 100 mg v 0.4 mm Hg with placebo; P less than 0.005). It also found that sildenafil significantly decreased diastolic blood pressure in men with cervical and thoracic-injured men (12 men with cervical spine injury and erectile dysfunction; mean change in supine diastolic blood pressure at 1 hour: 9.5 mm Hg with sildenafil 50 mg v 1.6 mm Hg with placebo; P less than 0.005; 11 men with thoracic spine injury and erectile dysfunction; mean change in supine diastolic blood pressure at 1 hour: 9.5 mm Hg with sildenafil 100 mg v 0 mm Hg with placebo; P less than 0.05). Dose dependant dizziness was also reported in all spinal injured men.

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Phosphodiesterase inhibitors

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men after radical prostatectomy or with prostate cancer: Sildenafil is more effective at improving the proportion of men with successful erections and successful intercourse ( moderate-quality evidence ). ADVERSE EFFECTS Sildenafil has been associated with headache, flushing, visual disturbances and dyspepsia. Sildenafil is contraindicated in men using oral nitrates concomitantly, in whom deaths have been reported.

Benefits

Sildenafil versus placebo in men after radical prostatectomy or with prostate cancer:

We found one systematic review (search date 2000, 27 RCTs) The systematic review included a subgroup analysis of men with a history of radical prostatectomy. It found that compared with placebo, sildenafil significantly improved the proportion of men with successful erections and successful intercourse (116 men with radical prostatectomy or prostate cancer and erectile dysfunction; AR for improved erections: 48% with sildenafil v 10% with placebo; RR 3.8, 95% CI 1.6 to 9.5; 42 men, AR for mean percentage of successful intercourse attempts: 25% with sildenafil v 3% with placebo; WMD 24, 95% CI 5 to 43). We found one small subsequent RCT in men with erectile dysfunction after external beam radiotherapy for prostate cancer. It found that compared with placebo, sildenafil significantly improved global efficacy and the proportion of men with successful intercourse after 6 weeks of treatment (60 men with radical prostatectomy or prostate cancer and erectile dysfunction; AR for global efficacy: 45% with sildenafil v 8% with placebo; P < 0.001; AR for successful intercourse: 55% with sildenafil v 18% with placebo; P less than 0.001).

Harms

The systematic review found that in a subset of 14 flexible dose trials (3780 men), sildenafil significantly increased the risk of at least one adverse effect compared with placebo (3780 men with any cause of erectile dysfunction; AR: 48% with sildenafil v 36% with placebo; RR 1.4, 95% CI 1.3 to 1.6). Adverse effects included headache (11% with sildenafil v 4% with placebo; RR 2.6, 95% CI 1.8 to 3.7), flushing (12% with sildenafil v 2% with placebo; RR 5.8, 95% CI 3.4 to 10), dyspepsia (5% with sildenafil v 1% with placebo; RR 3.8, 95% CI 2.2 to 6.6), and visual disturbance (3.0% with sildenafil v 0.8% with placebo; RR 3.1, 95% CI 1.8 to 5.4). Similar proportions of those allocated to sildenafil or placebo discontinued treatment because of adverse effects (1.3% with sildenafil v 1.2% with placebo; RR 1.3, 95% CI 0.7 to 2.3). The data from fixed dose trials in this systematic review indicate that all these adverse effects were more frequent at higher doses of sildenafil and were mild to moderate in severity.

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Vardenafil

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men after prostatectomy: Vardenafil may be more effective at improving erections and attempts at successful intercourse at 6 months in men with nerve sparing radical prostatectomy ( low-quality evidence ). ADVERSE EFFECTS Vardenafil has been associated with headache, flushing, and dyspepsia. Vardenafil is contraindicated in people receiving nitrates because of the risk of potentially life threatening hypotension.

Benefits

Vardenafil versus placebo in men after prostatectomy:

The systematic review identified one RCT comparing vardenafil versus placebo in men after prostatectomy. It found that, compared with placebo, vardenafil 10 or 20 mg significantly improved erections and successful attempts at intercourse as assessed by scores on the erectile function domain of the IIEF, questions 2 and 3 of the Sexual Encounter Profile diaries and the Global Efficacy Question after 12 weeks (440 men with erectile dysfunction [> 67% severe] at least 6 months after nerve sparing radical retro-pubic prostatectomy; P less than 0.0001 for all comparisons v placebo). Erectile function improved for men on vardenafil with all levels of severity of erectile dysfunction (absolute figures not reported).

Harms

Vardenafil versus placebo in men after prostatectomy:

The systematic review found that adverse effects with vardenafil 5–40 mg were reported in 22–61% of people in four RCTs. In the largest RCT (762 men in safety analysis), headache was the most common (10–21%) followed by rhinitis (9–17%), flushing (5–13%), and dyspepsia (1–6%). The rates of adverse events with placebo were not reported in the systematic review. The review also included a crossover, single dose study (41 men with stable angina) to assess exercise tolerance in men on vardenafil 10 mg or placebo. Vardenafil did not reduce symptom limited exercise time or time to awareness of angina compared with placebo (P = 0.39–0.59). Vardenafil use did increase time to ST segment depression compared with placebo (381 seconds with vardenafil v 334 seconds with placebo; P = 0.0004).

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Alprostadil (intraurethral)

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men with erectile dysfunction of any cause: Intraurethral alprostadil (prostaglandin E1) may be more effective at increasing the proportion of men achieving at least one successful attempt at sexual intercourse and at improving quality of life scores (relationship with partner, improvement in personal wellness and quality of erection) in men who have previously responded to alprostadil ( high-quality evidence ). Compared with placebo in men after radical prostatectomy: Intraurethral alprostadil is more effective at increasing the proportion of men reporting at least one successful attempt at sexual intercourse ( moderate-quality evidence ). Compared with intracavernosal alprostadil: Intraurethral alprostadil may be less effective at increasing the proportion of men who achieve an erection and successful intercourse ( very low-quality evidence ). ADVERSE EFFECTS Alprostadil has been associated with penile pain, urethral burning sensations and minor urethral trauma.

Benefits

Intraurethral alprostadil versus placebo in men with erectile dysfunction of any cause:

We found one systematic review (search date 2003, 3 RCTs, 1828 men, mostly organic causes for erectile dysfunction) that compared four doses of intraurethral alprostadil (125, 250, 500, and 1000 µg) versus placebo. Two of the RCTs were parallel group studies and one a crossover design. Combining the results of the two parallel group trials, the systematic review found that, compared with placebo, intraurethral alprostadil significantly increased the proportion of men achieving at least one successful attempt at sexual intercourse over a 3 month home self administration period (2 RCTs, 1101 men, AR for at least 1 successful attempt at intercourse: 345/528 [65.3%] with intraurethral alprostadil v 101/573 [17.6%] with placebo; OR 7.22, 95% CI 5.68 to 9.18). One of the parallel group RCTs identified by the systematic review found that, compared with placebo, alprostadil treated men reported an improvement in all three domains of an eight item quality of life questionnaire (improvement in relationship with partner: 34% with alprostadil v 11% with placebo; improvement in personal wellness: 5% with alprostadil v 8% with placebo; quality of erection: 71% with alprostadil v 1% with placebo; P less than 0.005 for each comparison). The partners of 35% of the men treated with alprostadil reported an improvement in their relationship compared with 12% of those treated with placebo (significance assessment not performed). The crossover RCT identified by the systematic review found that 63.6% of all men had “at least one successful intercourse” with at least one dose of intraurethral alprostadil (AR: 39.4% with alprostadil 125 µg v 33.3% with alprostadil 250 µg v 40% with alprostadil 500 µg v 50% with 1000 µg alprostadil v 12.5% with placebo; P less than 0.01 for each active dose compared with placebo). Alprostadil also increased the proportion of men with erections sufficient for intercourse compared with placebo (AR: 20% with alprostadil 125 µg v 30.3% with alprostadil 250 µg v 27% with alprostadil 500 µg v 32% with alprostadil 1000 µg v 5% with placebo; P less than 0.001 for all comparisons v placebo).

Intraurethral alprostadil in men after radical prostatectomy:

We found one RCT (270 men) that treated men with individually titrated doses of intraurethral alprostadil 125–1000 µg or placebo at home for 3 months. It found that intraurethral alprostadil increased the proportion of men reporting at least one successful intercourse compared with placebo (AR: 72/126 [57%] with alprostadil v 9/137 [7%] with placebo; P less than 0.001).

Intraurethral alprostadil versus intracavernosal alprostadil in men with erectile dysfunction of any cause:

We found three RCTs that compared intraurethral versus intracavernosal routes of administering alprostadil. The first RCT (111 men) compared home treatment with intracavernosal alprostadil (< 40 µg) versus intraurethral alprostadil (< 1000 µg) plus optional ACTIS (a penile constriction ring) for 4 weeks after an in-clinic dose titration period of 1–14 days. It found that intraurethral alprostadil reduced the proportion of men achieving at least one erection sufficient for intercourse compared with intracavernosal alprostadil (crossover open label design, mean age 59.2 years with any cause of erectile dysfunction, mean duration or 4.5 years; AR: 42/68 [62%] with intraurethral alprostadil v 63/68 [93%] with intracavernosal alprostadil; P less than 0.0001). It also found that intraurethral alprostadil reduced mean scores on questions 3 and 4 of the International Index of Erectile Function compared with intracavernosal treatment (mean score on the International Index of Erectile Dysfunction [range 0–5]; question 3: 1.7 at baseline and 3.0 at week 4 with intraurethral alprostadil v 1.7 at baseline and 4.4 at week 4 with intracavernosal alprostadil; question 4: 1.3 at baseline and 2.8 at week 4 with intraurethral alprostadil v 1.3 at baseline and 4.2 at week 4 with intracavernosal alprostadil; P less than 0.0001 for both comparisons). Fewer men and their partners expressed a preference for intraurethral alprostadil compared with intracavernosal alprostadil (AR for preference of mode of delivery of alprostadil; men: 16% with intraurethral alprostadil v 69% with intracavernosal alprostadil; partners: 10% with intraurethral alprostadil v 63% intracavernosal alprostadil; significance assessment not performed). The second RCT (60 men, organic causes of erectile dysfunction) compared intracavernosal alprostadil 20 µg with intraurethral alprostadil 1 mg. It found that intraurethral alprostadil reduced the proportion of men who reported erections sufficient for sexual intercourse and more than one successful intercourse compared with intracavernosal alprostadil (AR for successful erections: 60% with intraurethral alprostadil v 90% with intracavernosal alprostadil; AR for successful intercourse: 53% with intraurethral alprostadil v 87% with intracavernosal alprostadil; P less than 0.05 for both comparisons). A third RCT (103 men, mean age 51.7 years) compared intraurethral alprostadil (< 1000 µg) with intracavernosal alprostadil (< 20 µg). Intraurethral alprostadil reduced rates of successful erections compared with intracavernosal alprostadil (AR: 43% with intraurethral alprostadil v 70% with intracavernosal alprostadil; significance assessment not performed).

Harms

Intraurethral alprostadil versus placebo in men with erectile dysfunction of any cause:

The systematic review found that, compared with placebo, intraurethral alprostadil increased penile pain compared with placebo (2 RCTs, 1056 men; AR: 170/567 [30.0%] with alprostadil v 18/589 [3.1%] with placebo; OR 7.39 95% CI 5.40 to 10.12). One RCT also reported that men on intraurethral alprostadil reported penile pain significantly more frequently than those on placebo (AR: 38.9% with alprostadil v 1.5% with placebo; P < 0.001) as well as urethral burning (AR: 18.3% with alprostadil v 4.4% with placebo; P less than 0.001). The systematic review found that intraurethral alprostadil also increased the frequency of reports of minor urethral trauma compared with placebo (AR: 26/567 [4.6%] with alprostadil v 6/589 [1.0%] with placebo; OR 3.79, 95% CI 1.88 to 7.65). No significant difference in rates of urinary tract infection was reported in the systematic review (AR: 0.17% with alprostadil v 0.51% with placebo; OR 0.39, 95% CI 0.05 to 2.78). Priapism was reported by one person on intraurethral alprostadil (AR: 1/567 [0.18%] with alprostadil v 0/589 [0%] with placebo; OR 7.12, 95% CI 0.14 to 359.12). Dizziness was significantly more common with intraurethral alprostadil compared with placebo (AR: 11/567 [1.9%] with alprostadil v 1/589 [0.17%] with placebo; OR 5.57, 95% CI 1.79 to 17.37).

Intraurethral alprostadil versus intracavernosal alprostadil in men with erectile dysfunction of any cause:

One RCT comparing intracavernosal alprostadil versus intraurethral alprostadil found that penile pain was the most common adverse effect for both interventions (AR: 33.8% with intracavernosal alprostadil v 25.0% with intraurethral alprostadil). Local bleeding was reported in both intervention arms (AR: 1.5% with intracavernosal alprostadil v 2.9 with intraurethral alprostadil; significance assessment not performed). In another RCT, penile pain was significantly more common with intracavernosal alprostadil (AR: 7% with intraurethral alprostadil v 47% with intracavernosal alprostadil; P less than 0.05). However, the reverse was true in a third RCT (AR for penile pain/urethral burning: 31.4% with intraurethral alprostadil v 10.6% with intracavernosal alprostadil; significance assessment not performed). Temporary urethral bleeding was observed in 5/103 (4.8%) men after intraurethral alprostadil. Brief periods of dizziness (6.8%) and syncope (1 man) were also observed after intraurethral alprostadil, although this was not seen with the intracavernosal route of administration. We found no reports of penile fibrosis or other serious adverse events.

Comment

None of the RCTs described the randomisation or allocation concealment procedures. All the RCTs in the systematic review and two of the subsequent RCTs pre-selected men who had a good response to alprostadil before randomisation. This would tend to increase the size of the effect compared with placebo, affect the external validity of the results and consequently their generalisation to clinical practice.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Apomorphine

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men with erectile dysfunction of any cause: Sublingual apomorphine is more effective at increasing the proportion of men who achieve sufficient erections and successful intercourse ( moderate-quality evidence ). ADVERSE EFFECTS Apomorphine has been associated with nausea, dizziness, headache, drowsiness, and yawning. A drug safety alert has also been issued on the possibility of an association with pathological gambling and increased libido.

Benefits

We found no systematic review.

Apomorphine versus placebo in men with erectile dysfunction of any cause:

We found three RCTs that compared sublingual apomorphine with placebo. The first RCT found that, compared with placebo, sublingual apomorphine taken 15–25 minutes before attempted intercourse (2–4 mg escalated over 4 weeks and continued at 4 mg for 4 weeks) increased the proportion of people who achieved erections sufficient for intercourse and successful intercourse at 8 weeks (507 men aged 22–77 years, > 50% on concomitant medication; AR for successful erections: 62% with apomorphine v 55% with placebo; P = 0.02; AR for successful intercourse: 38% with apomorphine v 28% with placebo; P = 0.001). Sublingual apomorphine 2 mg (current recommended dose) only and 2–3 mg only, both significantly improved successful intercourse compared with placebo (AR: apomorphine 2 mg: figures not reported; P = 0.007; 2–3 mg apomorphine: 35% with apomorphine v 26% with placebo; P = 0.002). Escalation of the dose of apomorphine from 2 mg to 3 mg increased the proportion of men with successful intercourse compared with the 2 mg dose (P = 0.002), but the proportion of men with successful intercourse with escalation to the 4 mg dose was similar to 2–3 mg (figures not reported; significance assessment not performed). The second RCT compared sublingual apomorphine optimised to 2, 4, 5, or 6 mg over 4 weeks with the optimum dose continued for another 4 weeks versus fixed doses of apomorphine 5 or 6 mg for 8 weeks versus placebo for 8 weeks. All doses of apomorphine significantly increased the proportion of men with successful erections and the proportion of men with successful attempts at intercourse at 8 weeks (569 men aged 21–72 years, 32% with severe erectile dysfunction, mostly organic causes, crossover study design; AR for men with erections firm enough for intercourse: 53% with apomorphine 2, 4, 5 ,6 mg dose optimisation v 47% with apomorphine 2, 4 mg dose optimisation v 53% with apomorphine 5 mg fixed dose v 53% with apomorphine 6 mg fixed dose v 34% with placebo; P less than 0.002 for all comparisons v placebo; AR for proportion of men with successful intercourse attempts: 50% with apomorphine 2, 4, 5, 6 mg dose optimisation v 45% with apomorphine 2, 4 mg dose optimisation v 51% with apomorphine 5 mg fixed dose v 50% with apomorphine 6 mg fixed dose v 32% with placebo; P less than 0.01). The third crossover RCT had three study arms and compared sublingual apomorphine 3 and 4 mg versus placebo. Apomorphine 3 mg significantly increased the proportion of men with successful erections and proportion of men with successful attempts at intercourse compared with placebo (296 men aged 27–72 years, moderate to severe erectile dysfunction in > 50%, mostly organic causes; AR for successful erections: 48.3% with apomorphine 3 mg v 34% with placebo; P less than 0.001; AR for proportion of successful attempts at intercourse: 49% with apomorphine 3 mg v 35% with placebo). There was no significant difference between 3 mg and 4 mg of apomorphine in the proportion of men with successful erections and successful intercourse attempts (AR for men with successful erections: 49% with apomorphine 3 mg v 50% with apomorphine 4 mg; AR for proportion of men with successful attempts at intercourse: 48% with apomorphine 3 mg v 50% with apomorphine 4 mg; P greater than 0.40 for both comparisons). The advantage of apomorphine 3 mg over placebo was seen across all grades of severity of erectile dysfunction. The median time to erection with apomorphine 3 mg was 18.8 minutes (95% CI 17.0 minutes to 20.0 minutes). Apomorphine 3 mg significantly improved the proportion of men achieving erections firm enough for intercourse for mild, moderate, and severe cases of erectile dysfunction (proportion of successful erections in mild erectile dysfunction: 66% with apomorphine 3 mg v 51% with placebo; P = 0.004; moderate erectile dysfunction: 55% with apomorphine 3 mg v 35% with placebo; P = 0.002; severe erectile dysfunction: 33% with apomorphine 3 mg v 25% with placebo; P value not reported; significance assessment not performed).

Harms

Apomorphine versus placebo in men with erectile dysfunction of any cause:

Nausea, dizziness, somnolence, and headache were the most common adverse effects seen with sublingual apomorphine in all RCTs comparing apomorphine with placebo that reported adverse effects. One RCT found that, compared with placebo, apomorphine increased the proportion of men reporting nausea (25/224 [9.8%] with apomorphine v 1/224 [0.4%] with placebo; significance assessment not performed). Nausea was seen with the initial doses, was more common with 4 mg (13/25 [52%] men) and tended to subside with continued use. In addition, dizziness was reported by 7.1% of men with apomorphine and 2.4% with placebo, headache was reported by 6.7% with apomorphine and 4% with placebo. One RCT found that, compared with placebo, apomorphine 3 mg increased the proportion of men with nausea (7.0% with apomorphine v 1.1% with placebo; significance assessment not performed). This was the most common adverse effect. It also found that, compared with placebo, apomorphine 3 mg increased the proportion of men with yawning (8.1% with apomorphine v 0.6% with placebo), dizziness (6.5% with apomorphine v 3.4% with placebo), somnolence (4.9% with apomorphine v 1.7% with placebo), headache (2.2% with apomorphine v 4.5% with placebo), and vasodilatation (2.2% with apomorphine v 2.3% with placebo). One person on apomorphine 3 mg experienced vasovagal syncope after their first dose, although physical examination, vital signs and laboratory tests showed no clinically significant changes from baseline. The proportion of men with each adverse event observed was smaller with apomorphine 3 mg compared with apomorphine 4 mg (nausea: 3.3% with apomorphine 3 mg v 14.1% with apomorphine 4 mg; yawning: 8.9% with apomorphine 3 mg v 13% with apomorphine 4 mg; dizziness: 2.2% with apomorphine 3 mg v 5.4% with apomorphine 4 mg; somnolence: 5.6% with apomorphine 3 mg v 9.8% with apomorphine 4 mg; headache: 4.4% with apomorphine 3 mg v 6.5% with apomorphine 4 mg; vasodilatation: 1.1% with apomorphine 3 mg v 6.5% with apomorphine 4 mg). Similar dose dependant adverse effects were reported in another RCT. The incidence of these adverse events also dropped considerably over the last 4 weeks (out of 8 weeks) of treatment (proportion of men reporting nausea at 4 weeks: 24% with apomorphine 2, 4 mg dose optimisation v 34% with apomorphine 5 mg fixed dose v 44% with apomorphine 6 mg fixed dose v 3% with placebo; proportion of men reporting nausea at 8 weeks: 10% with apomorphine 2, 4 mg dose optimisation v 12% with apomorphine 5 mg fixed dose v 17% with apomorphine 6 mg fixed dose v 0% with placebo; significance assessment not performed). Another placebo controlled crossover RCT (162 men who were also on antihypertensive drugs, diuretics, and nitrates) specifically assessed the cardiovascular safety of apomorphine 5 mg administered on alternative days. It found no clinically important changes in heart rate or blood pressure in men taking common antihypertensives, and short acting nitrates, and most men on long acting nitrates with this higher than recommended dose of apomorphine. Men on long acting nitrates developed significant changes in standing blood pressure after sublingual apomorphine but not in supine blood pressure (mean standing systolic blood pressure change: –5 mm Hg to –9 mm Hg 30–60 minutes after treatment, mean standing diastolic blood pressure change: –3 mm Hg to –4 mm Hg 50–60 minutes after treatment; P < 0.05 for both comparisons with baseline readings). The rates of other adverse events after a single dose of apomorphine were greater than for placebo (AR for any adverse event: 21% with apomorphine v 8% with placebo; significance assessment not performed). The most common adverse events were dizziness (11% with apomorphine v 2% with placebo), nausea (10% with apomorphine v 1% with placebo), and headache (6% with apomorphine v 2% with placebo). Syncope occurred in one man on beta-blockers and symptomatic hypotension in eight men (1 each on angiotensin converting enzyme inhibitors, alpha1 blockers, beta blockers, diuretics, and 2 each on short and long acting nitrates).

Drug safety alert

MHRA issues drug safety alert on pathological gambling and increased libido possibly associated with apomorphine (7 November 2006)

Since the last update of this topic, a drug safety alert has been issued on pathological gambling and increased libido possibly associated with apomorphine (www.mhra.gov.uk).

Comment

None of the RCTs described methods of randomisation or allocation concealment.

Clinical guide:

Apomorphine may be a safer oral treatment than sildenafil, tadalafil, or vardenafil in men with erectile dysfunction who are taking nitrates.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Alprostadil (topical)

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo: Alprostadil (prostaglandin E1) gel applied to the tip of the penis may be more effective at improving erections and successful intercourse in men with various causes of erectile dysfunction ( low-quality evidence ). ADVERSE EFFECTS Topical alprostadil has been associated with penile pain and erythema.

Benefits

We found no systematic review.

Topical alprostadil versus placebo in men with erectile dysfunction of any cause:

We found two small RCTs that compared alprostadil gel or cream applied topically to the tip of the penis versus placebo. The first small RCT (42 men with various causes of erectile dysfunction) found alprostadil cream significantly improved successful erections and sexual intercourse as assessed by questions 3 and 4 of the International Index of Erectile Function (IIEF) compared with placebo (improvement in IIEF question 3 erection score from baseline [range: 1.67–2.05]: 10.19 with alprostadil v 1.4 with placebo; P < 0.01; improvement in IIEF question 4 successful intercourse score from baseline [range: 1.29–1.65]: 1.45 with alprostadil v 0.14 with placebo; P < 0.01). Alprostadil cream also improved erections as assessed by the Global Assessment Questionnaire compared with placebo (42 men: 73.7% with alprostadil cream v 19.0% with placebo; P < 0.01) and increased the proportion of successful attempts at sexual intercourse (42 men: 68.4% with alprostadil cream v 19.1% with placebo; P < 0.01; absolute numbers not reported). The second small RCT (60 men with moderate to severe erectile dysfunction) compared alprostadil 1% in a gel formulation with SEPA 5% (soft enhancer of percutaneous absorption) versus placebo gel. It found that alprostadil gel significantly increased the proportion of men with erections judged sufficient for vaginal penetration (patient assessment of rigidity as ≥ 3 on a scale of 1–5, or an angle of ≥ 70 degrees from vertical axis measured by physician) compared with placebo (12/31 [38.9%] with alprostadil v 2/29 [6.9%] with placebo; P = 0.005).

Harms

Topical alprostadil versus placebo in men with erectile dysfunction of any cause:

The first RCT reported a higher proportion of men experiencing adverse events with alprostadil cream compared with placebo (30% with alprostadil cream v 4.8% with placebo; P < 0.01). Common adverse events included mild pain of the penis and urethra. In the second RCT, significantly more men had erythema with alprostadil than with placebo (absolute numbers not reported; P < 0.001). Other adverse effects included conjunctivitis (2/31 [6.5%]) and hypotension (1/31 [3.2%]). A further RCT of two phase II trials (303 men with moderate to severe erectile dysfunction aged 21–70 years, 90% with erectile dysfunction > 1 year) compared topical alprostadil cream 0.05, 0.1, 0.2, and 0.3 mg versus placebo. It found a higher incidence of adverse events in men using alprostadil gel compared with placebo (AR for ≤ 1 adverse event: 135/230 [58.7%] with alprostadil gel v 25/75 [33.3%] with placebo; significance assessment not performed). Although over 97% of adverse events were described as mild and lasting 60 minutes or less, more men on alprostadil withdrew from therapy because of adverse events compared with placebo (AR: 37/230 [16.1%] with alprostadil cream v 0/75 [0%] with placebo; significance assessment not performed). Events resulting in withdrawal suggested a dose relation and included urogenital pain (22/230 [10%] with alprostadil cream v 0/75 [0%] with placebo; significance assessment not performed), and hypotension (13/230 [5.7%] with alprostadil cream v 0/75 [0%] with placebo; significance assessment not performed). One man using alprostadil cream 200 µg developed a near syncopal episode lasting nearly 10 minutes. About 2% of partners reported mild, transient vaginal burning with alprostadil cream. We did not find any RCTs or observational studies of sufficient quality on penile fibrosis, prolonged erections, or other serious adverse events.

Comment

None of the RCTs described the methods used to generate randomisation codes and to conceal allocation to treatment arms.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Papaverine

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with papaverine plus phentolamine (bimix): Papaverine is less effective at increasing the proportion of men achieving full erections 20 minutes after injection and at increasing the proportion of men achieving successful intercourse on the day of the injection ( moderate-quality evidence ). Compared with intracavernosal alprostadil: Papaverine may be less effective at increasing erections ( very low-quality evidence ). ADVERSE EFFECTS Papaverine has been associated with tranisent burning pain, bruising, prolonged erections, fibrosis of the corpora cavernosa, and fibrous penile nodules. NOTE We found no direct information about whether papaverine is better than no active treatment.

Benefits

We found no systematic review.

Papaverine versus placebo in men with erectile dysfunction of any cause:

We found no RCTs.

Papaverine versus papaverine plus phentolamine (bimix) in men with erectile dysfunction of any cause:

We found one crossover RCT (40 men, any cause of erectile dysfunction for > I year aged 40–75 years) that compared intracavernosal papaverine 40 mg versus papaverine 20 mg plus phentolamine 0.5 mg diluted with normal saline to 5 mL (bimix). It found that bimix significantly increased the proportion of men with full erections compared with papaverine alone, as assessed by blinded observers 20 minutes after injection (AR: 11/40 [27%] with papaverine v 19/40 [48%] with bimix; P < 0.05). Bimix also significantly increased the proportion of men achieving successful intercourse on the day of the injection (AR: 5/40 [12%] with papaverine v 15/40 [38%] with bimix; P < 0.05).

Papaverine versus intracavernosal alprostadil:

See benefits of alprostadil (intracavernosal).

Papaverine versus papaverine plus phentolamine plus alprostadil (trimix) in men with any cause of erectile dysfunction:

We found no RCTs.

Harms

Papaverine versus placebo in men with any cause of erectile dysfunction:

We found no RCTs. In a cohort of 226 men self injecting with papaverine and followed up for 2 years, 8% developed haematomas, 10% developed priapism, 12% developed fibrosis of the corpora cavernosa, and 9% developed fibrous penile nodules (significance assessments not performed).

Papaverine versus papaverine plus phentolamine (bimix) in men with any cause of erectile dysfunction:

The RCT found that 11/40 (27.5%) men receiving papaverine and 7/40 (17.5%) men receiving bimix experienced burning pain in the shaft of the penis 30 seconds after the injection, which subsided within 2 minutes after injection (significance assessments not performed). Prolonged erection occurred in 1/40 (2.5%) of men receiving bimix.

Papaverine versus intracavernosal alprostadil in men with any cause of erectile dysfunction:

See harms of alprostadil (intracavernosal).

Papaverine plus phentolamine (bimix) in men with any cause of erectile dysfunction:

See harms of papaverine plus phentolamine (bimix).

Papaverine plus phentolamine plus alprostadil (trimix) in men with any cause of erectile dysfunction:

See harms of papaverine plus phentolamine plus alprostadil (trimix).

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Papaverine plus phentolamine (bimix)

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo: Intracavernosal papaverine injections plus phentolamine (bimix) may be more effective at increasing the proportion of men with erections sufficient for intercourse ( low-quality evidence ). Compared with papaverine: Papaverine plus phentolamine (bimix) is more effective at increasing the proportion of men achieving full erections 20 minutes after injection and at increasing the proportion of men achieving successful intercourse on the day of the injection ( moderate-quality evidence ). Compared with intracavernosal alprostadil: We don't know whether papaverine plus phentolamine (bimix) is more effective at increasing the proportion of men with successful erections ( very low-quality evidence ). ADVERSE EFFECTS Papaverine plus phentolamine (bimix) has been associated with transient pain and bruises at injection site, painless fibrous penile nodules, mild to moderate alteration in liver function, prolonged erections, and fibrosis of the corpora cavernosa. NOTE We found no clinically important results about papaverine plus phenotolamine (bimix) compared with other treatments in men with erectile dysfunction.

Benefits

We found no systematic review.

Papaverine plus phentolamine (bimix) versus placebo in men with any cause of erectile dysfunction:

We found one small crossover RCT (30 men, mean age 60.9 years, mean duration of erectile dysfunction 4.8 years, erectile dysfunction resulting from various causes) that compared intracavernosal injections of a combination of papaverine 30 mg plus phentolamine 1 mg (bimix) versus normal saline injections. It found that papaverine 30 mg plus phentolamine 1 mg increased the proportion of men with erections satisfactory for intercourse compared with normal saline (AR: 82.8% with bimix v 0% with normal saline; significance not reported).

Papaverine plus phentolamine (bimix) versus papaverine in men with any cause of erectile dysfunction:

See benefits of papaverine.

Papaverine plus phentolamine (bimix) versus intracavernosal alprostadil in men with any cause of erectile dysfunction:

See benefits of alprostadil (intracavernosal).

Papaverine plus phentolamine (bimix) versus other treatments in men with any cause of erectile dysfunction:

We found no RCTs.

Harms

A prospective cohort study (111 men) found painless fibrous nodules in 57% of men self injecting bimix over 12 months (results presented graphically; P = 0.005). It also found that higher frequencies of injection significantly increased the proportion of men who developed nodules compared with a lower frequency (nodules present in men with a mean of 51.3 injections v no nodules present in men with a mean of 20 injections; significance assessment not performed). Priapism did not occur during home treatment but was seen in 2/329 (0.6%) of physician administered injections. Fifty men had at least one liver function test after the start of bimix injections; 20/50 [40%] of those 50 men had at least one abnormality in liver function, mostly involving mild to moderate elevation in alkaline phosphatase and lactic dehydrogenase (significance assessment not performed). A retrospective cohort study (224 men) found that 5% developed haematomas, 7% developed priapism, 9% developed fibrosis of the corpora cavernosa, and 8% developed fibrous plaques after self injecting with bimix over 2 years.

Papaverine plus phentolamine (bimix) versus placebo:

One man in the RCT developed a prolonged erection that resolved spontaneously after 26 hours with no subsequent abnormality. Most men (number not reported) experienced various degrees of ecchymosis at the injection site and some reported mild pain.

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Papaverine plus phentolamine plus alprostadil (trimix)

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with intracavernosal alprostadil: A mixture of papaverine, phentolamine, and alprostadil (trimix) may not only be more effective at increasing positive erectile responses and erections in men who have previously failed to respond to bimix, but is also associated with lesser penile pain ( very low-quality evidence ). NOTE We found no direct information about whether papaverine plus phentolamine plus alprostadil (trimix) is better than no active treatment.

Benefits

We found no systematic review.

Papaverine plus phentolamine plus alprostadil (trimix) versus placebo in men with any cause of erectile dysfunction:

We found no RCTs.

Papaverine plus phentolamine plus alprostadil (trimix) versus intracavernosal alprostadil in men with any cause of erectile dysfunction:

See benefits of alprostadil (intracavernosal).

Papaverine plus phenotolamine plus alprostadil (trimix) versus papaverine in men with any cause of erectile dysfunction:

We found no RCTs.

Harms

Papaverine plus phentolamine plus alprostadil (trimix) versus placebo in men with any cause of erectile dysfunction:

We found no RCTs.

Papaverine plus phentolamine plus alprostadil (trimix) versus intracavernosal alprostadil in men with any cause of erectile dysfunction:

See harms of alprostadil (intracavernosal).

Comment

None.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Alprostadil (intracavernosal)

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo in men erectile dysfunction of any cause: Alprostadil may be more effective at increasing the proportion of clinical assessments of full rigidity and RigiScan assessment of 70% or greater rigidity for 10 minutes or longer and at increasing the number of erections judged sufficient to allow penetration ( very low-quality evidence ). Compared with intraurethral alprostadil: Intracavernosal alprostadil may be more effective at increasing the proportion of men who achieve an erection and successful intercourse (very low-quality evidence). Compared with papaverine: Intracavernosal alprostadil may be more effective at increasing erections (very low-quality evidence). Compared with papaverine plus phentolamine (bimix): We don't know whether intracavernosal alprostadil is more effective at increasing the proportion of men with successful erections (very low-quality evidence). Compared with alprostadil plus papaverine plus phentolamine (trimix): A mixture of alprostadil, papaverine and phentolamine (trimix) may be more effective at increasing positive erectile responses and erections in men who have previously failed to respond to bimix (very low-quality evidence). Compared with sildenafil: Intracavernosal alprostadil and sildenafil are equally effective at 4–9 months (average 6 months) at improving erections sufficient for successful intercourse ( moderate-quality evidence ). Compared with vaccum devices: We don't know whether intracavernosal alprostadil is more effective at achieving an erection but intracavernosal alprostadil may be more effective at increasing overall satisfaction scores and the ability to achieve an orgasm (very low-quality evidence). ADVERSE EFFECTS Alprostadil has been associated with penile pain over the injection site.

Benefits

Intracavernosal alprostadil versus placebo in men with erectile dysfunction of any cause:

We found two RCTs that compared alprostadil injections with placebo. The first RCT (296 men aged 21–74 years, excluding men with penile deformities, uncontrolled diabetes or hypertension, major mental illness, infectious diseases, or a history of priapism) compared injections of 2.5, 5, 10, and 20 µg of alprostadil with placebo. None of the men responded to placebo and all doses of alprostadil increased the proportion of clinical assessment of “full rigidity” (P < 0.01) and RigiScan assessment of 70% or greater rigidity for 10 minutes or longer (P < 0.001). It also found a higher proportion of men with a clinical response with larger doses (P ≤ 0.001), suggesting a dose–response relationship. Absolute numbers were not reported and results were presented graphically. The second RCT was a small crossover study (60 men, mean age 58 years, erectile dysfunction > 6 months) that compared alprostadil 30 µg versus bimix (papaverine 30 mg plus phentolamine mesylate 0.5 mg) versus placebo (isotonic saline). It found that alprostadil significantly increased the number of erections judged sufficient to allow penetration compared with placebo (successful erections: 50% with alprostadil v 0% with placebo; P < 0.001).

Intracavernosal alprostadil versus intraurethral alprostadil in men with erectile dysfunction of any cause:

See benefits of intraurethral alprostadil.

Intracavernosal alprostadil versus papaverine in men with erectile dysfunction of any cause:

We found three crossover RCTs that compared alprostadil injections with papaverine injections. The first crossover RCT (single blind, 205 men, mean age 57.5 years, various causes of erectile dysfunction) compared a single injection of low dose alprostadil (5 µg) with papaverine (18 mg, 0.6 mL of a 30 mg/mL solution). Alprostadil resulted in a significantly higher proportion of erections judged adequate for penetration 10–20 minutes after the injection compared with papaverine (AR for full erection: 34/129 [26%] with 5 µg alprostadil v 17/129 [13%] with papaverine; P < 0.03). There was no significant difference in the percentage of successful intercourse attempts on the same day or over the next 4 weeks between alprostadil and papaverine (men attempting intercourse same day: 24/129 [19%] with alprostadil v 15/129 [12%] with papaverine; P = 0.077; men experiencing successful intercourse within 4 weeks: 12/129 [9%] with alprostadil v 6/129 [5%] with papaverine; P = 0.61). The second small crossover RCT (54 men, mean age 57 years, with vascular cause of erectile dysfunction) compared intracavernosal injections of alprostadil 20 µg versus papaverine 60 mg. Alprostadil injections resulted in partial or complete erections in more men than with papaverine (AR: 46% with alprostadil v 14% with papaverine; P < 0.002; absolute numbers not reported). The third small crossover RCT (52 men, mean age 48.6 years, duration of erectile dysfunction 0.7–6.0 years, various causes of erectile dysfunction) compared alprostadil 20 µg with papaverine 30 mg. Alprostadil significantly increased the proportion of men with successful erections (penis erect with 90° angle to vertical body axis and duration ≥ 2 hours) compared with papaverine (AR: 42/52 [81%] with alprostadil v 33/52 [63%] with papaverine; P = 0.01). In a subgroup of 24 men with suspected vascular cause of erectile disorder, alprostadil also significantly improved successful erections compared with papaverine (AR: 16/24 [67%] with alprostadil v 11/24 [46%] with papaverine; P < 0.04).

Intracavernosal alprostadil versus papaverine plus phentolamine (bimix) in men with erectile dysfunction of any cause:

We found two crossover RCTs that compared alprostadil versus bimix. The first small crossover RCT (60 men, mean age 58 years, erectile dysfunction > 6 months) compared alprostadil 30 µg versus bimix (papaverine 30 mg plus phentolamine mesylate 0.5 mg) versus placebo (isotonic saline). It found no significant difference between alprostadil and bimix in the proportion of men with successful erections (30/60 [50%] with alprostadil v 34/60 [56%] with bimix; P > 0.05. The second crossover RCT (91 men, mean age 55 years) compared alprostadil (20 µg) versus bimix (papaverine 30 mg plus phentolamine 1 mg) versus trimix (alprostadil 10 µg plus papaverine 15 mg plus phentolamine 0.5 mg). Penile rigidity was assessed by the same observer subjectively and objectively using callipers. The comparison between alprostadil and bimix revealed that both were equally effective in producing erections, with no significant difference in the mean total percentage of rigidity (60% with alprostadil v 59% with bimix; P > 0.46) or the proportion of men with a positive erectile response of rigidity of 60% or greater (58/82 [71%] with alprostadil v 46/82 [59%] with bimix; P > 0.12).

Intracavernosal alprostadil versus alprostadil plus papaverine plus phentolamine (trimix) in men with erectile dysfunction of any cause:

We found two crossover RCTs that compared alprostadil versus trimix. One small crossover RCT (32 men, mean age 61.3 years, erectile dysfunction > 6 months; all had failed to respond to 2 tests with papaverine plus phentolamine [bimix]) compared alprostadil 40 µg with trimix (1 mL solution containing alprostadil 5.8 µg/mL plus papaverine 17.64 mg/mL plus phentolamine 0.58 mg/mL). Trimix significantly increased the proportion of men who gained an erection sufficient to allow penetration compared with alprostadil alone (AR: 7/32 [22%] with alprostadil v 16/32 [50%] with trimix; P < 0.05). The second crossover RCT (91 men, mean age 55 years) compared alprostadil (20 µg) versus bimix (papaverine 30 mg plus phentolamine 1 mg) versus trimix (papaverine 15 mg plus phentolamine 0.5 mg plus alprostadil 10 µg). Trimix significantly increased the mean total percentage of rigidity of 60% or greater (60% with alprostadil v 66% with trimix; P = 0.0115) and the proportion of men with a positive erectile response of at least 60% rigidity (58/82 [71%] with alprostadil v 67/82 [82%] with trimix; P = 0.007).

Intracavernosal alprostadil versus sildenafil in men with erectile dysfunction of any cause:

See benefits of sildenafil.

Intracavernosal alprostadil versus vacuum devices in men with erectile dysfunction of any cause:

One small crossover RCT (50 men with erectile dysfunction, 44 of whom completed the study, mean age 62.3 years, erectile dysfunction > 6 months) looked at satisfaction levels among men and their partners using intracavernosal self injection of alprostadil compared with external vacuum devices over 18–24 months. It found no significant difference in ability to achieve an erection suitable for intercourse between alprostadil intracavernosal self injection compared with external vacuum device (mean self rated erectile quality on a scale of 1–10: 5.1 with alprostadil v 4.3 with vacuum device; reported as not significant). However, the ability to attain orgasm was significantly better with alprostadil (mean self rated penile sensation on a scale of 1–10: 5.2 with alprostadil v 4.5 with vacuum device; P < 0.05). Overall satisfaction scores were significantly higher for men and their partners using alprostadil compared with vacuum device (men's mean overall satisfaction on a scale of 0–10: 6.5 with alprostadil v 5.4 with vacuum device; P < 0.05; partners' mean overall satisfaction on a scale of 0–10]: 6.5 with alprostadil v 5.1 with vacuum device; P < 0.05).

Harms

Intracavernosal alprostadil versus placebo in men with erectile dysfunction of any cause:

In the subsequent multicentre RCT adverse effects with alprostadil included penile pain (54/237 [23%]) and priapism (5/237 [2.1%]). A cohort of 208 men self injecting with alprostadil followed up over 3 years, reported the following adverse effects: haematomas (0.5%), priapism (1.5%), fibrosis of the corpora cavernosa (1.0%), and fibrous penile nodules (0.5%).

Intracavernosal alprostadil versus papaverine in men with erectile dysfunction of any cause:

Penile pain from injection rather than needle insertion was reported with both treatments (pain on injection: 11/72 [15.3%] with alprostadil v 6/72 [8.3%] with papaverine; significance assessment not performed) in the first RCT. One man (1/72) on papaverine developed priapism, which resolved spontaneously within 8 hours. Mild pain on the injection site was reported in the second RCT (44% of men on papaverine v 45% with alprostadil; absolute numbers not reported) in the second RCT, along with dizziness and headache in two men on papaverine and one man on alprostadil. In the third RCT, 6/52 (11.5%) of men given alprostadil and 13/52 (25%) of men given papaverine reported transient, tolerable burning at the injection site. There were no reports of priapism, even among eight men taking alprostadil who had previously developed priapism with papaverine.

Intracavernosal alprostadil versus papaverine plus phentolamine (bimix) in men with erectile dysfunction of any cause:

The RCT found that penile pain was more frequent with alprostadil (21/60 [35%] with alprostadil v 9/60 [15%] with bimix; P < 0.05).

Intracavernosal alprostadil versus alprostadil plus papaverine plus phentolamine (trimix) in men with erectile dysfunction of any cause:

The RCT found that penile pain was more frequent with alprostadil (13/32 [41%] with alprostadil v 4/32 [12.5%] with trimix; P < 0.05). We found no long term data on adverse effects.

Intracavernosal alprostadil versus vacuum devices in men with erectile dysfunction of any cause:

No significant difference was noted in the frequency of adverse events between vacuum devices and alprostadil.

Comment

None of the RCTs described the randomisation or allocation concealment procedures. The primary outcome in this trial was assessed by the subjective impression of an observer who was not blinded to treatment assignment. In the RCT that compared injections and vacuum devices, 80% of the 44 couples who completed the study were still using one or the other treatment after 18–24 months.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Penile prostheses

Summary

We found no direct information about penile prostheses in men with erectile dysfunction. There is consensus that penile prostheses are likely to be beneficial. Mechanical failure and infections are the most serious complications of penile prosthesis implantation. Use of penile prostheses is usually considered only after less invasive treatments have failed.

Benefits

We found no systematic review or RCTs. For ethical reasons, RCTs of penile implants versus non-operative treatments for erectile dysfunction are unlikely to be carried out. Anecdotal evidence suggests that patient satisfaction may be high, but we found no studies of adequate quality to assess this. However, there is consensus belief that penile prostheses are likely to be beneficial.

Harms

We found one prospective cohort study (331 men with erectile dysfunction implanted with different types of penile prosthesis) with 10 year follow up. Adverse effects of surgery included postoperative wound infection (19/331 [5.7%]), most of which involved abscesses requiring surgery for prosthesis removal. Additional complications included pain lasting over 1 week (20/331 [6.0%]), local swelling lasting more than 1 month (18/331 [5.4%]), temporary haematoma of the penis and scrotum (16/331 [4.8%]), appearance of deformity requiring revision (9/331 [2.7%]), and inconvenience in daily life requiring removal of prosthesis (3/331 [0.9%]). Mechanical failure of the prosthesis was another complication of implantation (22/331 [6.6%]).

Comment

Clinical guide:

Use of penile prostheses is usually considered only after less invasive treatments have failed.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Vacuum devices

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with intracavernosal alprostadil: We don't know how vacuum devices and intracavernosal alprostadil compare at achieving an erection, but intracavernosal alprostadil may be more effective at increasing overall satisfaction scores and the ability to achieve an orgasm ( very low-quality evidence ). Vacuum devices plus psychosexual therapy compared with psychosexual therapy: We don't know whether a vacuum device plus psychosexual therapy (package of relationship therapy and modified Masters & Johnson sex therapy) is more effective at increasing the proportion of men assessed by a therapist as improved (very low-quality evidence). ADVERSE EFFECTS Vacuum devices have been associated with haematoma and blocked ejaculations, NOTE We found no direct information about whether vacuum devices are better than no active treatment.

Benefits

Vacuum devices versus placebo:

We found no systematic review or RCTs.

Vacuum devices versus intracavernosal alprostadil:

See benefits of alprostadil (intracavernosal).

Vacuum devices versus vacuum devices plus psychosexual therapy:

We found one RCT (45 couples) that compared psychosexual therapy (a package of relationship therapy and modified Masters & Johnson sex therapy; sessions were 45–50 minutes and once in 2 weeks for a minimum of 3 sessions; therapists were supervised) with a vacuum device in addition to psychosexual therapy after the second session. It found no significant difference in the proportion of couples assessed by their therapist as showing moderate to good improvement (study outcome was the therapist's impression of improvement, categorised as: very good, moderate, slight, and no improvement or worse) between a combination of a vacuum device plus psychosexual therapy and psychosexual therapy alone, although more couples having combined treatment were assessed as improved (AR: 84% with combination therapy v 60% with psychosexual therapy alone; absolute benefit increase +24%, 95% CI –2% to +47%; P = 0.096). The RCT may have been too small to detect a clinically important difference between groups.

Harms

Vacuum devices versus placebo:

We found no RCTs.

Vacuum devices versus intracavernosal alprostadil:

See harms of alprostadil (intracavernosal).

Vacuum devices versus vacuum devices plus psychosexual therapy:

Some people in the RCT reported that vacuum devices interfered with the spontaneity of intercourse owing to the long set up time, and that they sometimes caused insufficient erections owing to pivoting at the base of the penis (absolute numbers not reported). The RCT found that vacuum devices increased bruising compared with intracavernosal trimix, although this difference did not reach significance (16% with vacuum device v 9% with intracavernosal trimix). The RCT also noted that vacuum devices caused blocked ejaculations (9/23 [39%]), haematoma at the site of the constriction band (3/23 [13%]), and discomfort (12/23 [52%]).

Comment

The RCT comparing vacuum devices and psychosexual therapy was not blinded and this may have biased outcomes in favour of the combination treatment. The outcomes did not report specifically on erections or successful sexual intercourse. The RCT comparing intracavernosal alprostadil with vacuum devices used outcome assessments that were not validated (see comment of alprostadil (intracavernosal)).

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Psychosexual counselling

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with waiting list control: We don't know whether psychosexual counselling is more effective at improving sexual functioning in men with psychological erectile dysfunction but psychosexual counselling may be more effective at improving aspects of sexual attitudes and behaviours ( very low-quality evidence ). Compared with interpersonal therapy: We don't know whether psychosexual counselling is more effective at improving sexual functioning or sexual satisfaction at 15 weeks in men with psychological erectile dysfunction. Interpersonal therapy aimed at improving social skills may be more effective than psychosexual counselling at increasing the proportion of men who no longer have erectile dysfunction at 6–12 months and when combined with psychosexual counselling may be more effective than either treatment alone ( low-quality evidence ). Compared with vacuum devices plus psychosexual therapy: We don't know whether psychosexual therapy (package of relationship therapy and modified Masters & Johnson sex therapy) is more effective at increasing the proportion of men assessed by a therapist as improved (very low-quality evidence).

Benefits

We found no systematic review.

Psychosexual counselling versus a waiting list control:

We found two RCTs that treated single men with erectile dysfunction compared with a waiting list control group. One small RCT (21 men with psychological dysfunction) found that, compared with waiting list control, psychosexual counselling provided in groups significantly improved aspects of sexual attitudes and behaviours (satisfaction with overall sexual self image: results presented graphically; P < 0.01). It found no significant difference in the frequency of at least one successful intercourse between psychosexual counselling and waiting list control (results presented graphically; P < 0.10). The other RCT (69 men, excluded those with an organic basis for erectile dysfunction) compared four interventions; standard psychosexual counselling (masturbatory training to attain anxiety free erections, instructions in “sensate focus”, “squeeze”, and “stop–start” techniques to attain ejaculatory control, sensory awareness exercises to reduce performance anxiety and enhance erotic pleasure, correction of misconceptions regarding sexuality and relationships, and reducing fears of disclosure of sexual inadequacy) versus interpersonal difficulties oriented therapy (individually tailored social skills training) versus a combination of psychosexual counselling and interpersonal difficulties oriented training versus a 15 week waiting list control group. At the end of the 15 week treatment period, the combined scores for all forms of psychological treatments improved sexual activities (P < 0.03) and sexual satisfaction (P < 0.004) compared with scores of the men in the waiting list control group, who had not made clinically meaningful gains as assessed by the 258 item Derogatis Sexual Functioning Inventory. Sexual functioning and satisfaction did not differ in men assigned to the three psychological treatments.

Psychosexual counselling versus interpersonal therapy:

The RCT described earlier found no significant difference between psychosexual counselling alone, interpersonal therapy alone, or combination therapy in sexual functioning or sexual satisfaction at the end of 15 weeks of treatment. However, over the 6 and 12 month follow up, the proportion of men who no longer had erectile dysfunction was significantly greater for those treated with interpersonal therapy than those treated with psychosexual counselling (AR for men not meeting DSM III criteria for erectile dysfunction at 1 year: 78% with interpersonal therapy v 40% with psychosexual counselling; P < 0.02). Combination treatment also increased the proportion of men who no longer had erectile dysfunction at 6 month and 1 year follow up compared with either interpersonal therapy or psychosexual counselling alone (P < 0.03 for combination treatment v interpersonal therapy alone post-treatment and v psychosexual therapy alone at 6 months and 1 year; results presented graphically).

Psychosexual counselling versus vacuum devices:

See benefits of vacuum devices.

Harms

The RCTs gave no information on adverse effects.

Comment

None of the RCTs described randomisation or allocation concealment. One of the RCTs that compared psychosexual counselling with a waiting list control was too small to exclude a beneficial effect for psychosexual counselling.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Cognitive behavioural therapy

Summary

We found no direct information about cognitive behavioural therapy in men with erectile dysfunction. Anecdotal evidence suggests that this may be an effective treatment.

Benefits

We found no systematic review or RCTs.

Harms

We found no RCTs.

Comment

Clinical guide:

Cognitive behavioural therapy involves attempts to elicit and modify maladaptive thoughts and to address relationship issues, in addition to behavioural exercises (as in standard sex therapy). Modern sex therapy incorporates some aspects of cognitive behavioural therapy and conversely, cognitive behavioural therapy includes many elements of sex therapy. Anecdotal evidence suggests that it may be an effective treatment but we found no studies of adequate quality to assess this.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Ginseng

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo: Korean red ginseng seems to be more effective at increasing successful erections and sexual intercourse, and at improving sexual desire and intercourse satisfaction scores in men with mixed causes of erectile dysfunction ( moderate-quality evidence ).

Benefits

We found no systematic review.

Ginseng versus placebo:

We found one small crossover RCT (45 men, mean age 54 years, mixed causes of erectile dysfunction, 40–50% with severe erectile dysfunction) that compared Korean red ginseng versus placebo both taken three times daily over two 8 week periods separated by a 2 week washout period. It found that ginseng improved scores on the erectile function, sexual desire, and intercourse satisfaction domains of the International Index of Erectile Function compared with placebo (mean erectile function score: 11.24 with placebo v 15.02 with ginseng; mean sexual desire score: 4.40 with placebo v 5.28 with ginseng; mean intercourse satisfaction score: 5.56 with placebo v 7.10 with ginseng; P < 0.05 for all comparisons v placebo). Ginseng also increased successful erections and successful sexual intercourse as assessed by questions 3 and 4 on the International Index of Erectile Function compared with placebo (mean question 3 score [range 0–5]: 2.06 with placebo v 2.70 with ginseng; mean question 4 score: 1.92 with placebo v 2.83 with ginseng; P < 0.01 for both comparisons). Ginseng was rated by a larger proportion of men as improving erections as assessed by the Global Efficacy Question compared with placebo (AR for men with satisfactory erections: 60% with ginseng v 20% with placebo; P < 0.01).

Harms

The RCTs did not report on harms.

Comment

Clinical guide:

Ginseng is a traditional Asian remedy with rare adverse effects in the recommended dose of 0.5–2.0 g daily.

Substantive changes

No new evidence

BMJ Clin Evid. 2006 Mar 1;2006:1803.

Yohimbine

Summary

IMPROVEMENT IN SEXUAL FUNCTION Compared with placebo: Yohimbine may be more effective at improving self reported sexual function and penile rigidity at 2–10 weeks ( very low-quality evidence ).

Benefits

Yohimbine versus placebo:

We found one systematic review (search date 1997, 7 RCTs [including 5 crossover trials], 419 men with various causes of erectile dysfunction) comparing yohimbine versus placebo and one subsequent RCT. Meta-analysis found that yohimbine significantly improved erectile response (measurement varied between studies and included objective and subjective assessments of penile rigidity and sexual function) compared with placebo (erectile response: 34–73% with yohimbine v 9–45% with placebo; OR 3.85, 95% CI 2.22 to 6.67; absolute numbers not reported). The subsequent RCT, a small crossover study (29 men, mixed types of erectile dysfunction, mean age 51 years) compared yohimbine 36 mg daily with placebo over two 25 day treatment periods with a 14 day washout period in between. It found no significant difference in positive clinical responses between yohimbine and placebo (positive clinical response: 12/27 [44%] with yohimbine v 13/27 [48%] with placebo; reported as not significant).

Harms

The review reported that adverse effects were generally minor and reversible, and included agitation, anxiety, headache, mild increase in blood pressure, increased urinary output, and gastrointestinal upset. Yohimbine was associated with a higher proportion of adverse events compared with placebo (10–30% with yohimbine v 5–16% with placebo; significance assessment not performed). Two men withdrew from the subsequent crossover RCT because of a hypertensive crisis in one and severe palpitations in the other.

Comment

None of the RCTs in the systematic review or the subsequent RCT described the method of randomisation. The trials included in the systematic review were clinically heterogeneous in aetiology of erectile dysfunction and age, and lacked homogenous study designs (combined partial crossover and parallel RCTs) or homogenous outcome assessments. The forest plot of the meta-analysis did not indicate heterogeneity, but statistical heterogeneity was not assessed.

Substantive changes

No new evidence

Articles from BMJ Clinical Evidence are provided here courtesy of BMJ Publishing Group

RESOURCES