The introduction of oral agents for the treatment of erectile dysfunction (ED) has revolutionized the treatment of men with erection problems of all severities and etiologies. Sildenafil, available on the world market since 1998, was joined in 2003 by tadalafil and vardenafil as effective safe and reliable oral agents for the treatment of ED. While these agents have the same mechanism of action, there are differences among the three agents. Sildenafil has the longest patient experience and the most robust data confirming its activity, safety, and tolerability. It has recently been released for use in pulmonary hypertension as well as ED.

Among the earliest recognized phosphodiesterase (PDE) inhibitors were caffeine and theophylline. The first PDE inhibitor used clinically for the treatment of erectile dysfunction was papaverine, a non-selective inhibitor of PDE-5. Papaverine is administered via injection into the corpora cavernosa of the penis, either alone or in combination with other vasoactive agents. Sildenafil, the first oral PDE-5 inhibitor, was approved for use in men with erectile dysfunction in 1998. The addition of vardenafil and tadalafil to the market has increased the number of approved PDE-5 inhibitors to three agents used throughout the world. Each of these agents has a similar mechanism of action, but there are pharmacological and clinical differences.

All three approved agents in this class have similar pharmacokinetic and pharmacodynamic profiles and each is effective for all ages of patients with erectile dysfunction of all severities and etiologies. While there are clear pharmacokinetic and pharmacodynamic differences amongst these agents, clinical differences are somewhat more difficult to identify. Indeed the data of preference trials, head-to-head clinical trials, and selection trials are few. The differences in pharmacokinetics, while having distinct advantages in marketing each drug, may be difficult for clinicians and patients to identify. With the lack of data and good-quality clinical trials, it is difficult for the clinician to differentiate among the three agents and to select a PDE-5 inhibitor for a specific erectile dysfunction patient or a specific agent to switch to if an initial PDE-5 agent is unsuccessful or poorly tolerated.



One classification of differences amongst the three PDE-5 inhibitors is that of specificity for phosphodiesterase inhibition. Because the PDE enzyme is a participant in erectile function, inhibition is a method of facilitating erectile function. cGMP facilities the relaxation of smooth muscle cells in the corpus cavernosum through the nitric oxide (NO) pathway. Because PDE-5 in the corpus cavernosum smooth muscle cells breaks down cGMP, inhibition or blockade of this enzyme will prolong the duration and increase the concentration of cGMP in the smooth muscle cell, facilitating erectile function. Currently 11 families of PDEs have been identified in the human. Selectivity of the three currently available PDE-5 agents is predominantly for PDE-5; however, there is some additional inhibition of other PDE enzymes by the various agents. Sildenafil has excellent selectivity for PDE-5 over all the other PDEs except for PDE-6, which has some degree of inhibition from both sildenafil and vardenafil, most significantly sildenafil. This selectivity for PDE-6 produces a dose-related impairment of blue–green color discrimination and leads to the blue vision that some patients report. This PDE-6 inhibition is not related to the blindness caused by non-arteritic ischemic optic neuropathy reported with some PDE-5 agents.

Absorption and metabolism

Because of differences in gastrointestinal absorption with fatty meals and time of absorption, the three drugs have differences in ultimate peak plasma concentrations based on food intake. Sildenafil, when taken with a high-fat meal, has a reduction in maximum concentration (C max) of approximately 29 % , with delays in time to peak plasma concentration (T max) that can be as long as 1 hour. This interaction may result in delayed onset or reduced efficacy because of a decrease in peak serum concentration. This reduction may result in treatment failure in some patients. Thus, patients taking sildenafil should be instructed to do so either 1–2 hours after eating or with a reduced fat meal.

The cytochrome P (CYP) 450 system is the chief metabolic pathway for sildenafil, which is a CYP3A4 substrates. Similarly, sildenafil has a desmethyl metabolite that accounts for approximately 20 % of its overall pharmacologic activity. Sildenafil has a terminal half-life of approximately 4–5 hours. Excretion is principally as fecal metabolites, with 80 % of oral doses excreted in the feces.

In men ages 65 years or older, the area under the curve (AUC, i.e. the systemic exposure) of sildenafil increases by 40%. Therefore, lower starting doses are recommended for the elderly, beginning at 25 mg with dose escalation as required by patient response and side-effects.

Duration of action

The three available PDE-5 inhibitors have some significant differences in serum half-life. Sildenafil has a half-life of 4 hours, but it may be clinically active for as long as 8 hours because of prolonged adherence to receptor sites beyond the usual serum half-life. The clinician must be aware that emergency treatment with nitrate medications following ingestion of sildenafil should be delayed for 24 hours.

Onset of action

Multiple studies have been performed to evaluate the onset of activity of the three PDE-5 inhibitors. Because of the artificial nature of these studies, clinical relevance continues to be controversial. Onset of action studies are designed to use stopwatch evaluations by patients or partners following ingestion of the PDE-5 agent. Significance in onset of action is measured as first measurement of statistically significant difference in erectile function compared with placebo. While this statistically significant difference may occur as early as 11 minutes with sildenafil, success at this early time occurs for fewer than 40 % of patients treated. In patients with significant risk factors and comorbidities for ED, counseling to begin sexual activity at 15 minutes or earlier leads to treatment failure and may, in the final analysis, create performance anxiety, and patients may inappropriately lose confidence in the treatment.

Drug interactions

Use of any PDE-5 inhibitor agent is absolutely contraindicated in patients taking any form of nitric acid (NO) donor, especially organic nitrates. Sildenafil can potentiate the vasodilator and hypotensive effects of these agents. Reductions in blood pressure with nitrates have been observed within 24 hours after taking sildenafil; therefore, nitrates should not be used to manage acute myocardial ischemia in a patient who has taken sildenafil within the prior 24 hours.

Acute postural hypotensive symptoms have also been observed when sildenafil is administered together with the alpha-adrenergic blocker doxazosin, which is indicated for the treatment of benign prostatic hyperplasia as well as hypertension. Sildenafil at doses exceeding 25 mg should not be administered within 4 hours of a patient taking an alpha-blocker.

Concomitant administration of sildenafil with other anti-hypertensive agents (e.g. calcium-channel blockers) does not result in significant declines in blood pressure or in increased occurrence of hypotension, syncope, or other adverse cardiovascular events compared with placebo.

Alcohol is a mild vasodilator with effects on NO synthase activity and NO output by endothelial cells. In a placebo-controlled crossover trial involving eight healthy non-smoking men, there were no untoward hemodynamic interactions between sildenafil 100 mg and red wine (750 ml, 13.5 % v/v) consumed 60 minutes after sildenafil dosing (i.e. time of peak plasma concentration for sildenafil). Red wine intake significantly elevated cardiac index and heart rate, and sildenafil alone significantly lowered mean arterial pressure by up to 7 % and peripheral vascular resistance by up to 8 %. However, the combination had no effect on any of these parameters compared with red wine alone.

Potent CYP3A4 inhibitors, including HIV protease inhibitors (e.g. indinavir, ritonavir), azole antifungals, and mac-rolide antibiotics (e.g. erythromycin, azithromycin) can increase systemic exposure (the AUC) of sildenafil by two- to 16-fold. Conversely, CYP3A4 inducers such as rifampin can reduce circulating PDE-5 inhibitor levels.

Grapefruit juice inhibits first-pass CYP metabolism in the gastrointestinal tract and may thus increase oral bioavail-ability of the PDE-5 inhibitors. A randomized crossover trial evaluated the effects of taking a single sildenafil 50 mg dose 1 hour before, or at the same time as, 250 ml of grapefruit juice in 24 healthy male volunteers. Compared with water (reference period), grapefruit juice consumption increased the AUC values for sildenafil and N-desmethylsildenafil by about 23-24 % and slightly prolonged the time to peak plasma concentration (by about 15 minutes). Grapefruit juice also rendered sildenafil pharmacokinetics more variable, and the authors concluded that the combination should be avoided, especially in patients who might be prone to more marked hemodynamic effects.

Adverse events

Most adverse events (AEs) associated with sildenafil therapy are caused by inhibition of PDE-5 in tissues other than the corpus cavernosum. Sildenafil-associated AEs occur in >3 % of patients and include:

  • • headache in 16 % of patients taking sildenafil compared with 4 % of men receiving placebo;
  • • flushing (10% vs 1%);
  • • dyspepsia (7% vs 2%); and
  • • nasal congestion (4% vs 2%).

Discontinuations because of AEs were infrequent, occurring in 2.5 % of patients receiving sildenafil compared with 2.3 % of patients receiving placebo. AEs appear to decline over time in published studies of sildenafil.


Extensive treatment of patients with vascular risk factors using sildenafil has demonstrated safety in cardiac patients. Indeed, sildenafil was originally designed as a cardioprotective agent, and clinical and laboratory studies have confirmed this safety. The Princeton Consensus Guideline Conference II carefully reviewed the risks, AEs, and safety of PDE-5 inhibitors in men with cardiac disease. This expert conference, with meta-analysis of available phase III studies, demonstrated no increased risk of cardiac events in patients taking PDE-5 inhibitors compared with placebo-treated patients or patients in the general, age-adjusted population with similar age and risk factor profiles. Indeed, in several of the studies reviewed, patients taking regular PDE-5 inhibitors were demonstrated to have fewer cardiac events than those not taking PDE-5 inhibitors.

While clinical trials of sildenafil have included large numbers of patients with cardiovascular disease and diabetes mellitus, they have excluded patients with unstable cardiovascular disease. Accordingly, PDE-5 inhibitors are either not recommended or are to be used with caution in men with unstable angina, recent myocardial infarction, cardiac failure, a life-threatening or uncontrolled arrhythmia, poorly controlled blood pressure (resting blood pressure less than 90/50 mmHg or more than 170/100–110 mmHg), or heart failure. In addition, patients with left ventricular outflow obstruction secondary to aortic stenosis or idiopathic hypertrophic subaortic stenosis, as well as men with severe autonomic insufficiency, may be especially sensitive to the vasodilator effects of PDE-5 inhibitors. Consensus guidelines have been issued for risk-stratifying and counseling patients with concomitant sexual dysfunction and cardiovascular disease. Men with erectile dysfunction and associated significant cardiovascular conditions must be counseled that there is a transient increase in the relative risk of cardiovascular events during and within about 2 hours after sexual intercourse with or without PDE-5 inhibitor treatment.

Exercise echocardiographic studies have demonstrated that treatment with sildenafil 50–100 mg did not adversely affect hemodynamic variables, exercise capacity (treadmill time), or time to cardiac ischemia (or first awareness of angina) in patients with stable coronary artery disease.

In clinical trials of sildenafil, the incidence rate of myocardial infarction ranged from 1.0 per 100 patient-years for patients taking sildenafil in open-label studies to 1.7 per 100 patient-years for those receiving sildenafil in randomized controlled trials, as against 1.4 per 100 patient-years for those randomized to placebo in such trials. The slight disparities between incidence rates of myocardial infarction across sildenafil trials may reflect differences in age or other patient characteristics at baseline.

Even at supratherapeutic doses consistent with concomitant treatment with CYP inhibitors or renal impairment, sildenafil does not increase the corrected QT interval in a clinically significant manner. Sildenafil 50–400 mg does not increase the absolute QT interval, and elicited only mild rises in the Fridericia-corrected QT interval duration 1 hour after dosing in healthy adult males ages 45–60 years (mean, 53) who also received the active control moxifloxacin. These findings, together with the absence of reports of torsades de pointes in sildenafil post-marketing surveillance databases, suggest that none of the PDE-5 inhibitors causes clinically significant prolongation of the QT interval.

Case reports have implicated PDE-5 inhibitors in non-arteritic ischemic optic neuropathy (NAION), a rare form of acute blindness usually found in elderly patients with significant vascular risk factors. While fewer than 50 cases have been reported in the world literature, multiple clinical, epidemiologic, and basic science studies have been completed to examine the link between PDE-5 inhibitors and blindness. This association is not related to the blue-tinted vision reported by some sildenafil patients. This latter finding is transient and related to the inhibition of PDE-6 in the cones of the eye. Animal studies of the optic nerve after sildenafil demonstrate increased optic nerve blood flow, opposite to the expected finding with non-arteritic ischemic optic neuropathy. Similarly, an estimate of the frequency of cases in a non-treated population shows similar numbers of cases to that reported in association with PDE-5 use, with no signal to increased prevalence.

The systemic exposures of sildenafil may be increased in patients with renal insufficiency or hepatic impairment. Starting or other doses may need to be limited to sildenafil 25 mg in patients with these conditions.

Clinical effectiveness data

The US Food and Drug Administration approved sildenafil in March 1998 after extensive clinical trials. Because sildenafil has an almost 10-year history of safety and efficacy, it has more extensive clinical and laboratory data to demonstrate its use in varied etiologies, populations, and severities of ED. Early data for sildenafil include a 12-week randomized, placebo-controlled study of sildenafil 50 mg that demonstrated 65 % of successful intercourse attempts compared with 20 % in a placebo group. In reviewing these studies, however, it is clear that 25–35 % of patients in these clinical study protocols had inadequate responses to PDE-5 inhibitors. Sildenafil phase 3 trials were reported in the New England Journal of Medicine in 1998. In a dose-escalation (50–100 mg) study involving 329 men (mean age, 59 years) with erectile dysfunction for about 5 years (organic erectile dysfunction in 55 % or more of cases), the mean score for the erectile function domain of the International Index of Erectile Function (IIEF) at the end of 12 weeks of treatment was 22.1 (mild) in the sildenafil group compared with 12.2 (moderate) in the placebo group (p < 0.001). Scores on the orgasmic function, intercourse satisfaction, and overall satisfaction domains (but not the sexual desire domain) also significantly improved.

Long-term efficacy

With the increasing experience of PDE-5 inhibitor treatment for ED, efficacy over time is becoming better documented. Long-term studies including 5- and 6-year data from sildenafil have demonstrated no clinical evidence for tachyphylaxis. Indeed, the long-standing efficacy of sildenafil, the agent longest on the market, strongly suggests that this class of agents continues to be effective with long-term use. Since these agents are used only on an as-needed basis, few patients have taken daily doses for long periods of time. The few long-term daily dosing studies, however, have not demonstrated conclusively any evidence for tolerability or tachyphylaxis. Similarly, daily dosing studies, while still limited in duration, show improved, not compromised, responses in comparison with on-demand dosing.

Difficult-to-treat patients

All three PDE-5 inhibitors have been demonstrated to be effective in patients with severe erectile dysfunction. Indeed, patients with prostate cancer who have had radiation therapy or radical surgery and patients with severe vascular disease, diabetes, or depression are all treated satisfactorily with these agents. While the efficacy declines in these patients with severe erectile dysfunction, these agents are safe and effective, but consideration in these patients should be given to increasing dosage levels to maximum acceptable dose.

Men with diabetes mellitus are at elevated erectile dysfunction risk and are difficult to treat. Sildenafil has been demonstrated as effective therapy for men with diabetic ED. The Sildenafil Diabetes Study Group reported that 56 % of men with diabetic erectile dysfunction who received sildenafil (25–100 mg) treatment for 12 weeks reported improved erections, in contrast to 10 % of patients receiving placebo (p < 0.001). In this study, 61 % of men randomized to sildenafil reported at least one successful attempt at sexual intercourse compared with 22 % of controls (p < 0.001).

Because intact innervation to the penis is necessary for physiologic erectile responses, substantial proportions of patients with prostate cancer experience erectile dysfunction following either nerve-sparing radical retropubic prostatectomy or radiation therapy. Prostate cancer patients treated with sildenafil have shown significant improvements in erectile function. In an open-label sildenafil study involving 84 men (mean age, 62 years) with erectile dysfunction 2.1 years post-prostatectomy, 53 % of patients receiving treatment at doses of 50–100 mg reported improved erections, and 40 % reported an enhanced ability to achieve and maintain erections. Erectile function was directly related to the degree of nerve sparing, with patients having had a bilateral nerve-sparing procedure tending to respond better than those having had a unilateral or non-nerve-sparing procedure. Lower pathological stage and older patient age were also predictive of improved outcomes.

Using a single question form the Erectile Dysfunction Inventory of Treatment Satisfaction rather than the entire instrument, Hong et al. documented a sildenafil treatment satisfaction rate of only 26 % between 0 and 6 months after nerve-sparing radical retropubic prostatectomy, which rose to a maximum of 60 % between the 18th and 24th postoperative month.

Failure of phosphodiesterase type 5 inhibitor therapy

Because as many as 30–40 % of patients will not respond to PDE-5 inhibitors alone, strategies must be considered to enhance responses. Most importantly, patients must be counseled in the proper administration of PDE-5 inhibitors. For sildenafil, patients should be counseled to avoid high-fat meals, and patients with significant comorbidities should be advised to delay sexual stimulation for 1 hour following administration. Similarly, patients should be counseled that sexual stimulation is necessary. Dose escalation is also critical in achieving therapeutic success. The majority of men in marketing studies have optimal doses of 100 mg. Studies with sildenafil have demonstrated an improvement in response after patients have taken sildenafil six or more times. Because many patients have had prolonged ED, they should be counseled that multiple doses may be necessary before optimum response is achieved. Patients who continue to be poorly responsive to PDE-5 inhibitors should be further evaluated for hypogonadism. Indeed, sildenafil has been demonstrated to function poorly in the presence of low testosterone levels. Normalizing testosterone with testosterone gel therapy and maximizing the sildenafil dose to 100 mg will increase responses substantially. Indeed Shabsigh et al. showed improvement in erectile function domain scores of the IIEF of 4.4 points in patients treated with testosterone and sildenafil, compared with sildenafil 2.1 points with testosterone and placebo, a statistically significant difference. Additionally, those patients treated with the combination therapy had an improvement in ejaculatory function. Doses of sildenafil above the recommended 100 mg may be effective in some men in whom standard doses fail. McMahon showed a 24 % success rate using 200 mg after previous sildenafil failures. Drop-outs were 31 % and side-effects increased in all categories.

Future directions

PDE-5 inhibitors were introduced in an effort to improve erectile function. Their efficacy and safety have been well recorded in millions of patients worldwide. These agents, however, should not be confined to the treatment of ED. Early data support the improvement of endothelial cell-mediated flow in peripheral arteries via the NO pathways, suggesting a possible use for these agents as treatment for conditions known to limit endothelial function. Sildenafil is approved in the USA and in Europe for the treatment of pulmonary hypertension. Trials have confirmed the improved pulmonary dynamics with sildenafil, which appears to be more effective than standard treatment, with improvements in lifestyle and functional status.

In addition, data have demonstrated the efficacy of chronic dosing of PDE-5 inhibitors in patients following radical prostatectomy. Rehabilitation using sildenafil was demonstrated to improve post-radical prostatectomy erectile function by seven-fold in a study performed by Padma-Nathan et al.

Sildenafil is an effective and safe PDE-5 inhibitor, and the clinician now has multiple choices in treatment of patients with erectile dysfunction of all severities and etiologies. Since no well-controlled, head-to-head selection or patient preference studies are available, each clinician must choose an agent based on the profile of the patient, his tolerance, risk factors, and side-effects. Patients in whom activity is limited because of cardiac disease should be evaluated before PDE-5 inhibitors are prescribed, and no PDE-5 inhibitor should be prescribed in patients taking nitrate medications.


Selections from the book: “Textbook of Erectile Dysfunction”, 2009

Share →