Ian Eardley MA MChir FRCS(Urol) FEBU, Department of Urology, St James University Hospital Leeds, UK. Tel: 44 113 2066994; Fax: 44 113 2064920; E-mail: firstname.lastname@example.org
Introduction. Pharmacotherapy is the usual initial therapy for most men with erectile dysfunction.
Aim. To review the current data relating to the efficacy, tolerability and safety of drugs used in the treatment of men with erectile dysfunction.
Methods. A critical review of the literature relating to the use of pharmacotherapeutic agents was undertaken by a committee of eight experts from five countries, building on prior reviews.
Main Outcome Measures. Expert opinion and recommendations were based on grading of evidence-based literature, internal committee dialogue, open presentation, and debate.
Results. Almost all currently available evidence relates to sildenafil, tadalafil, and vardenafil. Phosphodiesterase type 5 (PDE5) inhibitors are first-line therapy for most men with erectile dysfunction who do not have a specific contraindication to their use. There is no evidence of significant differences in efficacy, safety, and tolerability between the PDE5 inhibitors and apomorphine. Intracavernosal injection therapy with alprostadil should be offered to patients as second line therapy for erectile dysfunction. Intraurethral alprostadil is a less effective treatment than intracavernosal alprostadil for the treatment of men with erectile dysfunction.
Conclusions. PDE5 inhibitors are effective, safe, and well-tolerated therapies for the treatment of men with erectile dysfunction. Apomorphine, intracavernosal injection therapy with alprostadil, and intraurethral alprostadil are all effective and well-tolerated treatments for men with erectile dysfunction. We recommend some standardization of the assessment of psychosocial outcomes within clinical trials in the field of erectile dysfunction. Eardley I, Donatucci C, Corbin J, El-Meliegy A, Hatzimouratidis K, McVary K, Munarriz R, and Lee SW. Pharmacotherapy for erectile dysfunction. J Sex Med 2010;7:524–540.
For most men with erectile dysfunction (ED), pharmacotherapy provides the initial form of therapy. Historically, locally active agents were used, such as injectable papaverine, phentolamine, and prostaglandin. In the late 1990s, the first effective oral agent, namely sildenafil, became available and it was soon followed by other similar drugs such as vardenafil and tadalafil. While all these agents acted peripherally within the penis, drugs that work within the central nervous system to improve erectile function have proven more difficult to develop, with apomorphine currently the only licensed agent.
This article summarizes the major findings of the 3rd International Consultation relevant to the pharmacotherapy of ED. For each agent, we outline the purported mechanism(s) of action and relevant basic science before moving on to an evidence-based review of the clinical data.
Two groups of drugs are worthy of mention. The first group contains drugs that have been reviewed in previous consultations but which are not widely used at this time, and for which there are little (if any) new data (notably yohimbine and trazodone). For these drugs, we have not included any comments. The second group of drugs includes those that are currently under clinical development. Inevitably, there is less published data relating to these drugs and this inevitably restricts our ability to make significant recommendations regarding these drugs.
Overall, we are able to make the following general recommendations:
Drugs which are effective when administered systemically
In most countries, the phosphodiesterase inhibitors are the initial form of pharmacotherapy for men with ED. They are orally active agents that are typically taken, on demand, prior to intercourse. Three such drugs, sildenafil, tadalafil, and vardenafil, are licensed for use around the world. A number of newer agents are available in a few countries (udenafil, mirodenafil) and others (avanafil, lodenafil, SLX-2101) are currently under development.
The normal pathway for penile erection is initiated by sexual arousal, which stimulates release of nitric oxide at nerve endings in the penis (Figure 1). Another source of nitric oxide is vascular endothelial cells. Nitric oxide diffuses into vascular smooth muscle cells in the penile corpus cavernosum to cause stimulation of guanylyl cyclase and elevation of cyclic guanosine monophosphate (cGMP) in these cells. This leads to activation of cGMP-dependent protein kinase (PKG), phosphorylation of several proteins, and lowering of cell calcium or reduction in sensitivity to calcium, which results in smooth muscle relaxation. The increased accumulation of blood in corpus cavernosum caused by this relaxation is the underlying basis for penile erection. Lack of proper cGMP elevation could be due at least in part to insufficient release of nitric oxide from nerve endings or endothelium. PDE5 inhibitors enhance erectile function during sexual stimulation by penetrating into smooth muscle cells and inhibiting PDE5, which is an enzyme that degrades cGMP. This results in decreased degradation of cGMP, which maintains higher cellular levels of cGMP in both corpus cavernosum and the vessels supplying it. This increases relaxation of the smooth muscle, which dilates the corporeal sinusoids resulting in increased blood flow, allowing an erection to occur. The pathway shown in Figure 1 may not work properly if the cGMP level in corpus cavernosum smooth muscle cells is not elevated sufficiently or if relaxation of smooth muscle in the tissue is deficient/incomplete [1,2].
PDE5 was discovered by Corbin and colleagues [3,4]. A cartoon depicting the PDE5 structure is shown in Figure 2. The enzyme is a homodimer containing two identical subunits with molecular weight of about 100,000 daltons per subunit. Each of the two subunits has a catalytic domain and a regulatory domain. The catalytic domain is the target of PDE5 inhibitors and it contains a single binding site for cGMP. When cGMP occupies this site the catalytic machinery, which is located very near the catalytic cGMP-binding site, breaks the cyclic phosphate bond of cGMP to form linear 5′-GMP. This dampens or terminates cGMP action. Because they have similar structures as cGMP, sildenafil, or other PDE5 inhibitors can also occupy the catalytic site, thus blocking access to cGMP. In fact, sildenafil occupies the site about 1,000 times more avidly than does the natural substrate, cGMP. However, the catalytic machinery does not break down the PDE5 inhibitors. Occupation of the catalytic site by these inhibitors competitively inhibits cGMP breakdown as cGMP cannot bind to gain access to the catalytic machinery. Inhibition of cGMP breakdown leads to elevation of cGMP in smooth muscle cells of the penile corpus cavernosum, resulting in relaxation of the muscle and penile erection.
Major research efforts have led to the production and development of compounds that are selective and potent in inhibiting particular PDEs . Some of these inhibitors have a similar overall structure as sildenafil and some, such as tadalafil, have significantly different structures (Figure 3). A part of the structure of each PDE5 inhibitor also resembles the structure of cGMP (see circled component). This is important since these drugs are competitive inhibitors (antagonists) of cGMP for PDE5, and they are believed to form some of the same molecular interactions as cGMP forms with amino acids in PDE5.
Assuming all other factors are equal, the higher the affinity (potency) of a PDE5 inhibitor for PDE5, the lower the expected dose of the inhibitor that will be needed . This concept of potency can be assessed by measuring the concentration of a particular PDE5 inhibitor in vitro that inhibits PDE5 activity by 50%, and is known as IC50 (Table 1). Highly potent drugs are expected to have affinities (IC50 values) in the nanomolar (nM) range. Vardenafil, SLX-2101 (Surface Logix, Inc., Brighton, MA, USA), and mirodenafil are within a class of biochemically high-affinity PDE5 inhibitors as compared with sildenafil, tadalafil, and udenafil, and avanafil has an intermediate potency. However, as discussed further, factors such as pharmacokinetics have strong impact on the dose required. Higher potency does not mean that a PDE5 inhibitor has a greater clinical effect, but that less of it is needed for the desired effect.
Table 1. Pharmacodynamics of the phosphodiesterase type 5 (PDE5) inhibitors
The biochemical selectivity of an inhibitor for PDE5 is a key factor in determining its side-effect profile . Once a large enough separation exists between the affinity (IC50) of the inhibitor for PDE5 and its affinity for nontarget PDEs (or other proteins), the less likely it is that it can achieve sufficient plasma concentrations to activate the nontarget site at therapeutic doses. For PDE5 inhibitors, selectivity is usually expressed in terms of potency (IC50) to inhibit PDE5 as opposed to inhibiting any others in the PDE family. PDEs are comprised of 11 families of enzymes that catalyze the termination of second messenger activity in cells by breaking the phosphodiester bond of either cAMP, cGMP, or both. These families (PDE1 to PDE11) are known or implicated in a broad range of cellular functions. PDE5 is present in high concentrations in the smooth muscle of corpora cavernosa of the penis. Sildenafil and vardenafil cross-react slightly with PDE6, i.e., their IC50s for PDE5 are only four- to 10-fold lower than those for PDE6. This may explain the complaint of some patients that sildenafil or vardenafil causes visual disturbances as PDE6 predominates in the retina. Tadalafil cross-reacts with PDE11 to some extent, but the consequences of this effect are unknown. PDE11 is found in the heart (cardiac myocytes), testes (germinal cells and Leydig cells), and anterior pituitary. Neither of the PDE5 inhibitors cross-reacts to a large extent with any of the other PDEs. Except for visual disturbances, the other reported side effects of PDE5 inhibitors (headaches, flushing, slight lowering of blood pressure, etc.) are likely caused by PDE5 inhibition in smooth muscle and other tissues outside the penile corpus cavernosum.
In addition to biochemical properties discussed above, pharmacokinetic properties of PDE5 inhibitors (ingestion/food interaction, movement in the circulation, tissue uptake, elimination) have great impact on efficacy . Several common pharmacokinetic parameters that can be measured and quantified describe distribution of a PDE5 inhibitor. The bioavailability, maximum plasma concentration (Cmax), time (tmax) required for attaining Cmax, and time (t1/2) required for elimination of one-half of the inhibitor from plasma are all important factors (Table 2). Sildenafil, vardenafil, udenafil, and avanafil have broadly similar tmax, which predicts a similar time of onset of action. The t1/2 values of tadalafil and udenafil are longer than those of the other two PDE5 inhibitors, which could be caused by the slower intestinal absorption and/or slower degradation of this drug by the liver, or it could be caused by other factors. The extended t1/2 of tadalafil provides a longer therapeutic effect , and this may also be the case for udenafil and SLx-2101. The Cmax of vardenafil is significantly lower than that for sildenafil and tadalafil, which might be expected based on its lower bioavailability.
Table 2. Pharmacokinetics of the phosphodiesterase type 5 inhibitors
PDE5 inhibitors are believed to be degraded in the liver. Therefore, as they are not degraded by PDE5 or any other enzyme in smooth muscle cells of corpus cavernosum, they must dissociate from PDE5, exit the smooth muscle cells, and then be transported to the liver via the bloodstream before they can be degraded (Figure 1). The rate of exit of a PDE5 inhibitor from smooth muscle cells should be considered when comparing PDE5 inhibitors as this could affect its duration of action. Disappearance of the inhibitor from plasma may imply, but does not prove, its disappearance, or clearance, from the cells in which it produces its effects. Since the inhibitor binds tightly to PDE5 in these cells, as portrayed in cartoon form in Figure 1, this could significantly retard its exit from these cells and prolong effects of PDE5 inhibitors in patients. It is conceivable that PDE5 inhibitors with higher affinities for PDE5 would dissociate from the enzyme more slowly, resulting in a more retarded clearance from corpus cavernosum cells. Studies of clearance of PDE5 inhibitors from plasma are documented, but studies of clearance of these inhibitors from smooth muscle cells are rare.
There is clear evidence that sildenafil is efficacious in the treatment of the broad population of men with erectile dysfunction when taken on demand at a doses of 25, 50, and 100 mg. There are multiple level 1 studies with consistent outcomes [20–43]. The overall grade of recommendation is grade A. In addition (again with a Grade A recommendation) there is clear evidence that sildenafil is efficacious in men where the ED has arisen as a consequence of specific diseases such as diabetes [44–47], depression [48–51], spinal cord injury [52,53], multiple sclerosis [54,55], cardiovascular disease [56–59], hypertension , and finally it is effective in men with lower urinary tract symptoms and ED .
Although side effects do occur with sildenafil (most notably headache, flushing, indigestion, nasal congestion, and occasional visual changes), providing that the drug is used in line with the labeling recommendations, there is no convincing evidence in the literature of any significant safety issue, including cardiovascular, visual, and aural safety [20–74] (Consistent level 1 studies, Overall grade of recommendation is grade A).
There is clear evidence that tadalafil is efficacious in the treatment of ED in the broad population when taken on demand at doses of 10 and 20 mg. There are multiple level 1 studies with consistent outcomes [75–88]. The overall grade of recommendation is grade A. In addition there are two level 1 studies that confirm that tadalafil is also efficacious when taken daily [89,90]. Although several doses have been tested, the doses that have been licensed are 2.5 and 5 mg. The overall grade of recommendation is grade A. Further, there is clear evidence that tadalafil is efficacious in a number of special populations of men where the ED has arisen as a consequence of a specific disease such as diabetes [91,92], men with ED who have undergone bilateral nerve sparing radical prostatectomy , men with ED who have undergone external beam radiotherapy for prostatic cancer , men with ED secondary to spinal cord injury  and men with lower urinary tract symptoms and ED [96,97].
Although side effects do occur with tadalafil (most notably headache, flushing, indigestion, nasal congestion, and back or girdle pain), providing that the drug is used in line with the labeling recommendations, there is no convincing evidence in the literature of any significant safety issue, including cardiovascular, visual, and aural safety (Consistent level 1 studies [75–104] with an overall grade of recommendation is grade A).
There is clear evidence that vardenafil is efficacious in the treatment of ED in the broad population at doses of 10 and 20 mg taken in an on-demand fashion. There are multiple level 1 studies with consistent outcomes [105–116]. The overall grade of recommendation is grade A. Further, there is clear evidence that vardenafil is efficacious in a number of special populations of men where the ED has arisen secondary to a specific disease such as diabetes [117–119], men with ED secondary to radical retropubic prostatectomy , depression , hypertension , spinal cord injury and ED , hyperlipidemia  and men who have previously failed to respond to sildenafil .
Although side effects do occur with vardenafil (most notably headache, flushing, indigestion and nasal congestion), providing that the drug is used in line with the labeling recommendations, there is no convincing evidence in the literature of any significant safety issue, including cardiovascular, visual and aural safety [105–127].
A single phase 3 study confirms that udenafil is effective in a broad population of Korean men with ED (level of evidence: 1) . The most common adverse events were facial flushing, nasal congestion, ocular hyperemia and headache. At the time of writing, much of the other data relating to udenafil is available in abstract form only and therefore not open to critical assessment.
Mirodenafil is a new PDE5 inhibitor already available in Korea (approved in November 2007). Preliminary results from a phase 2 clinical study  and a phase 3 clinical study  provided evidence for the efficacy and safety of mirodenafil (level of evidence: 1). The optimal doses in terms of efficacy and safety were determined from this study to be 50 mg and 100 mg. The most common adverse events were facial flushing, headache, nausea, and eye redness. At the time of writing, much of the other data relating to mirodenafil is available in abstract form only and therefore not open to critical assessment.
Lodenafil carbonate is another new PDE5 inhibitor under development in Brazil. It is a dimer formed by two lodenafil molecules linked by a carbonate bridge. After ingestion, the bridge is broken delivering the active compound lodenafil. Recently, a phase 2 clinical trial has been completed in Brazil (level of evidence: 2)  supporting efficacy and safety at doses of 20 mg, 40 mg and 80 mg. Adverse reactions included headache, dyspepsia, rhinitis, flushing and color visual disorders. At the time of writing, much of the other data relating to lodenafil is available in abstract form only and therefore not open to critical assessment.
Avanafil is a pyrimidine derivative synthesized as a PDE5 inhibitor. At the time of writing, all of the clinical data relating to avanafil is available in abstract form only and therefore not open to critical assessment, but doses of 50, 100, 200, and 300 mg have been investigated. Currently, phase 3 clinical trials are running to assess efficacy and safety of avanafil in general ED population and diabetic men with ED. The results are expected in early 2010.
SLx-2101 is a new PDE5 inhibitor that it is converted to an M1 metabolite, SLx-2081, which continues to be active. At the time of writing, all of the other data relating to SLx-2101 is available in abstract form only and therefore not open to critical assessment, but doses of 5, 10, 20, 40 and 80 mg have been investigated.
Centrally Acting Drugs
Dopamine is one among a number of important central neurotransmitters involved in the initiation of erection. It is the main transmitter within the PVN  and plays an important role in the central control of erection. Dopamine receptors are divided into two main families, D1 and D2-like receptors, which are in turn further subdivided into D1 to D5 receptor subtypes. Apomorphine has a higher affinity for the D2-like receptors  that are thought to be the main site for the induction of erections in the PVN . Apomorphine is therefore postulated to increase erectile responses by acting as a conditioner in the PVN, increasing the response to sexual stimuli resulting in enhanced erections induced in the periphery .
Overall there is evidence that apomorphine is efficacious in the treatment of erectile dysfunction in the broad population at doses of 2 and 3 mg taken sublingually in an on demand fashion. There are two level 1 studies [136,137] and two level 2 studies [138,139] with consistent outcomes. The overall grade of recommendation is grade A. There are several publications attesting to the tolerability of apomorphine at a dose of 2 and 3 mg [136–143]. The most common side effects are nausea, headache, and dizziness, with small numbers of patients developing syncope. This latter side effect was particularly noted at doses higher than those licensed for use in Europe. Overall, providing the drug is used in line with labeling we could find no evidence of significant tolerability of safety issues.
Bremelanotide (PT-141) is a synthetic peptide analogue of α-melanocyte stimulating hormone (α-MSH) activating melanocortin receptors 3 and 4 (MC3R and MC4R). It is an active metabolite of melanotan II. The effects of α-MSH on sexual behavior, including grooming, stretching, yawning, and penile erection, have been demonstrated in laboratory animals, and it is believed to act downstream from dopamine and oxytocin in hypothalamic centres near the third ventricle at the MC4R .
When injected subcutaneously it has been reported to initiate penile erection in normal men as well as men with psychogenic and organic erectile dysfunction [145–147]. It is also noted to significantly increase sexual desire. The frequent side effects associated with melanotan II administration are nausea and stretching/yawning. Although clinical trials have been undertaken to assess the potential use of an intranasal formulation of this compound for the treatment of men with erectile dysfunction [148,149], it has not yet been licensed, and at the time of writing, development of bremelanotide for men with ED appears to have ceased.
Comparator Studies of Systemically Active Agents
A number of studies have compared the efficacy and safety of the oral medications used in the treatment of men with ED. The trial designs used was variable with many studies demonstrating inadequate design resulting in clear bias. Almost all the studies were pharmaceutically sponsored.
The authors consider that one trial comparing sildenafil and tadalafil could be considered a Level 1 study [150–152] with the other published studies suffering from biases such as inadequate duration, inadequate washout, and biased dosing. The single level 1 study was an open-label study, but in other respects was well conducted and of adequate size. Overall, there was no evidence of superiority for any one of the PDE5 inhibitors over the others in terms of efficacy or safety (Grade A recommendation).
The authors consider that several trials comparing the efficacy and tolerability of apomorphine and sildenafil could be considered level 1 trials [153–156]. All were all open-label studies, but were otherwise well designed trials, with adequate numbers of patients randomized. Overall there was clear evidence that sildenafil has greater efficacy in the treatment of men with ED than apomorphine.
The dawn of the age of pharmacologic treatment began 25 years ago with the recognition that vasoactive drugs when injected into the penile erectile tissue were capable of initiating and maintaining erection [157,158]. Currently there are three main agents for intracavernosal injection (ICI) therapy one of which is approved by prominent national drug approval agencies (i.e., the Food and Drug Administration, European Agencies) for the treatment of ED. Additionally one agent for intraurethral therapy of ED has also received approval.
These were relegated to second line therapy after the appearance of effective oral PDE5 inhibitors. However, the local delivery of medications remain useful for the treatment of men with ED as PDE5 inhibitors are ineffective in about 25–32% of men others will ultimately fail to respond to oral therapy secondary to progression of their disease and a small number of men have contraindications to PDE5 inhibitors. The result is a large number of men who are unable to utilize oral treatments for ED.
PGE1-induced relaxation of human corpus cavernosum smooth muscle was first described by Karim and Adaikan (1975) . PGE1 mediates relaxation of corpus cavernosum smooth muscle via activation of EP prostaglandin receptors that results in an increase in the intracellular concentration of cAMP in corpus cavernosum smooth muscle [160–162]. PGE1 may also act by inhibiting the release of noradrenaline from sympathetic nerves  and suppressing angiotensin II secretion in the cavernosal tissues . PGE1 is metabolised by the 15-hydroxydehydrogenase present in the corpus cavernosum. The ability of human corpus cavernosum to degrade PGE1 probably aids in regulating the activity of PGE1 and reducing the risk of undesirable side effects such as prolonged erection and priapism.
Intracavernosal prostaglandin E1 is licensed for the treatment of men with ED and its efficacy was confirmed in three multi-centered, randomized prospective clinical trials with a six month open-label extension . Penile pain was the most commonly reported adverse event, occurring in 50% of the men at some time but was usually mild. Prolonged erections occurred in 5% of the men.
Vasoactive Intestinal Polypeptide and Phentolamine
Phentolamine is a nonselective alpha-adrenoceptor antagonist with similar affinity for alpha 1- and alpha 2 adrenoceptors. Phentolamine induced relaxation of corpus cavernosum erectile tissue is thought to occur by direct antagonism of alpha 1- and alpha 2-adrenoceptors, as well as by indirect functional antagonism via a non-adrenergic, endothelium-mediated mechanism suggesting nitric oxide synthase activation [166,167]. The clinical utility of phentolamine is presumably a reflection of the contribution of adrenergic neurotransmission to the maintained rugosity of the penis, and thus, inhibition of alpha-adrenoceptor activity alone may be sufficient for erection to commence . Oral/intracavernosal phentolamine therefore may facilitate penile erection by inhibiting the functional predominance of alpha 1-adrenoceptor activity that maintains erectile tissues in a non-erect state.
Vasoactive intestinal peptide (VIP) is a naturally occurring neurotransmitter. VIPergic nerves are most densely concentrated in the penis around the pudendal arteries and in the erectile tissue of the corpus cavernosum. VIP is known to exert regulatory actions on blood flow, secretion, and muscle tone. Its presence in considerable amounts in the male genital tract suggests that this peptide neurotransmitter may be important in the nervous control of male external genitalia . VIP co-localizes with NOS within the perivascular and trabecular nerve fibers innervating the penis . Most of these NO- and VIP-containing nerves appear to be cholinergic, as they also contain vesicular acetylcholine transporter, a specific marker for cholinergic neurons . The effects of VIP are mediated by a specific membrane-bound receptor linked to adenylate cyclase via a stimulatory G-protein. VIP has been shown to elevate cAMP concentrations in cavernosal tissues without affecting cGMP levels [172,173].
The first instance of the use of VIP as intracavernosal injection monotherapy for ED was disappointing  and it has shown more promise when used for combination therapy with phentolamine. Two multicentered prospective clinical trials using combination VIP/PH therapy have confirmed the efficacy and safety of the combination and have used to successfully register VIP as an approved pharmacologic agent for the treatment of men with ED in the United Kingdom, Denmark, and New Zealand [175,176]. Most commonly observed adverse effects were facial flushing and headache, characteristic events noted with vasoactive therapy.
Intracavernosal papaverine injection was the first clinically effective pharmacological therapy for ED. It is a smooth muscle relaxant. In vitro, papaverine evokes relaxation of isolated corpus cavernosum smooth strips, penile arteries, cavernous sinusoids, and the penile veins and attenuated contractions induced by stimulation of adrenergic nerves and exogenous noradrenaline [177,178]. It acts as a nonspecific phosphodiesterase inhibitor that initiates an increase in intracellular cAMP and cGMP leading to corporal smooth muscle relaxation and penile erection. Papaverine may also regulate cavernosal smooth muscle tone via inhibition of voltage-dependent L-type Ca2+ channels independent of cAMP as demonstrated in tracheal smooth muscle and suppression of angiotensin II secretion in cavernosal tissue .
Papaverine was the first agent discovered to be effective as intracavernosal pharmacotherapy for ED although it is not licensed for this use . It is effective but has fallen out of favor as monotherapy because of its high rates of fibrosis and priapism.
Phentolamine, papaverine, PGE1, and VIP are the vasoactive agents most commonly used in combination therapy to treat ED. Combination therapy is not only predictably more efficacious as a result of well-planned strategies based on sound pharmacological principles but it is also associated with a reduction in incidence of side effects and cost per dose.
In vitro studies on human and rabbit cavernosal strips demonstrated that phentolamine significantly potentiated relaxation induced by sildenafil, VIP, and PGE1. These vasodilators also significantly enhanced relaxation induced by phentolamine in the cavernosal tissue strips. The enhancement by phentolamine of VIP and PGE1-induced relaxation (cAMP-mediated) suggests a synergistic interaction whereas the interaction between phentolamine and sildenafil (cGMP-mediated) appears to be additive . The same investigators also show that sildenafil and PGE1 has additive and synergistic effect respectively with phentolamine-induced relaxation.
Although combination therapy remains unlicensed, there is intermediate level evidence for efficacy and safety [181,182].
Medicated urethral system for erection (MUSE) is a licensed alternative way to deliver alprostadil to the corporal bodies. MUSE involves the insertion of the delivery catheter into the meatus and depositing an alprostadil pellet in the urethra and the method is based on the absorption of drug through the urethral mucosa and into the corpora cavernosum. The efficacy and safety was confirmed in large controlled trial [183,184] with a systematic review of its use confirming its value . The commonest side effect is urethral pain. Comparator trials between MUSE alprostadil and ICI alprostadil, although of low quality, suggest significantly superior efficacy for intracavernosal alprostadil [186,187].
Although there has been considerable interest in the potential use of topical agents applied to the skin of the penis, as a treatment for ED, to date, the clinical results have been disappointing. In published trials of topical alprostadil with a variety of skin penetration enhancers, efficacy rates are low and at this time, such compounds remain investigational [188−192]. To date there are no significant publications that confirm efficacy for the topical application of agents such as papaverine, nitroglycerin, and minoxidil.
Pharmacotherapy remains the first-line therapy for men with ED. The evidence base for the currently licensed agents is strong, with the orally active PDE5 inhibitors being the first-line therapy for most men. Effective centrally acting agents with an adequate tolerability and safety profile have proven more difficult to develop. Locally active agents have a less robust evidence base to support efficacy, tolerability, and safety but remain an option when the systemically active agents have failed or are not appropriate. The principle conclusions and recommendations are shown in Table 3.
Table 3. Main recommendations
1. Phosphodiesterase type 5 (PDE5) inhibitors are effective, safe, and well-tolerated therapies for the treatment of men with erectile dysfunction. Almost all currently available evidence relates to sildenafil, tadalafil, and vardenafil (Grade A).
2. PDE5 inhibitors are first-line therapy for most men with erectile dysfunction who do not have a specific contraindication to their use (Grade C).
3. There is no evidence of significant differences in efficacy, safety, and tolerability between the PDE5 inhibitors (Grade A).
4. Apomorphine is an effective and well-tolerated treatment for men with erectile dysfunction (Grade A).
5. Sildenafil has superior efficacy to apomorphine in the treatment of men with erectile dysfunction (Grade A). The are no trials comparing the other PDE5 inhibitors and apomorphine
6. We recommend some standardization of the assessment of psychosocial outcomes within clinical trials in the field of erectile dysfunction (Grade C).
7. Intracavernosal injection therapy with alprostadil is an effective and well-tolerated treatment for men with erectile dysfunction (Grade A).
8. Intracavernosal injection therapy with alprostadil should be offered to patients as second-line therapy for erectile dysfunction (Grade C).
9. Intraurethral alprostadil is an effective and well-tolerated treatment for men with erectile dysfunction (Grade A).
10. Intraurethral alprostadil is a less effective treatment than intracavernosal alprostadil for the treatment of men with erectile dysfunction (Grade A).
Conflict of Interest: Dr. Eardley is a consultant, speaker and investigator for Pfizer, Bayer-Schering, Lilly, Boehringer Ingelheim. Dr. Hatzimouratidis is a speaker and investigator for Eli Lilly. No conflict of interest to declare for Drs. Corbin, El-Meliegy and Munarriz.