This AUA guideline is provided free of use to the general public for academic and research purposes. However, any person or company accessing AUA guidelines for promotional or commercial use must obtain a licensed copy. To obtain the licensable copy of this guideline, please contact Keith Price at kprice@auanet.org.
To cite this guideline:
Burnett AL, Nehra A, Breau RH et al: Erectile dysfunction: AUA guideline. J Urol 2018; 200: 633.
The purpose of this guideline is to provide a clinical strategy for the diagnosis and treatment of erectile dysfunction (ED).
Arthur L. Burnett, MD; Ajay Nehra, MD; Rodney H. Breau, MD; Daniel J. Culkin, MD; Martha M. Faraday, PhD; Lawrence S. Hakim, MD; Joel Heidelbaugh, MD; Mohit Khera, MD; Kevin T. McVary, MD; Martin M. Miner, MD; Christian J. Nelson, PhD; Hossein Sadeghi-Nejad, MD; Allen D. Seftel, MD; Alan W. Shindel, MD.
The Panel would like to dedicate this guideline to the memory of our friend and colleague, Ralph Alterowitz. We will forever be grateful to his contributions and devotion to the field of men's sexual health. He brought compassion and joy to all of those who were fortunate enough to work with him.
The purpose of this guideline is to provide a clinical strategy for the diagnosis and treatment of erectile dysfunction (ED). Using the shared decision-making process as a cornerstone for care, all patients should be informed of all treatment modalities that are not contraindicated, regardless of invasiveness or irreversibility, as potential first-line treatments. For each treatment, the clinician should ensure that the man and his partner have a full understanding of the benefits and risk/burdens associated with that choice.
The sexual response cycle is conceptualized as a sequential series of psychophysiological states that usually occur in an orderly progression. These phases were characterized by Masters and Johnson as desire, arousal, orgasm, and resolution. Erectile dysfunction (ED) can be conceptualized as an impairment in the arousal phase of sexual response and is defined as the consistent or recurrent inability to attain and/or maintain penile erection sufficient for sexual satisfaction, including satisfactory sexual performance. 1,2 The Panel believes that shared decision-making is the cornerstone of the treatment and management of ED, a model that relies on the concepts of autonomy and respect for persons in the clinical encounter. It is also a process in which the patient and the clinician together determine the best course of therapy based on a discussion of the risks, benefits and desired outcome. Using this approach, all men should be informed of all treatment options that are not medically contraindicated to determine the appropriate treatment. Although many men may choose to begin with the least invasive option, the Panel notes that it is valid for men to begin with any type of treatment, regardless of invasiveness or reversibility. Men also may choose to forego treatment. In each scenario, the clinician's role is to ensure that the man and his partner have a full understanding of the benefits and risks/burdens of the various management strategies.
A systematic review of the literature using the Pubmed, Embase, and Cochrane databases (search dates 1/1/1965 to 7/29/17) was conducted to identify peer-reviewed publications relevant to the diagnosis and treatment of ED. The review yielded an evidence base of 999 articles after application of inclusion/exclusion criteria. These publications were used to create the guideline statements. If sufficient evidence existed, then the body of evidence for a particular treatment was assigned a strength rating of A (high quality evidence; high certainty), B (moderate quality evidence; moderate certainty), or C (low quality evidence; low certainty). Evidence-based statements of Strong, Moderate, or Conditional Recommendation, which can be supported by any body of evidence strength, were developed based on the balance of benefits and risks/burdens to men and their partners. Additional information is provided as Clinical Principles and Expert Opinion when insufficient evidence existed.
Evaluation and Diagnosis:
Treatment:
6. For men being treated for ED, referral to a mental health professional should be considered to promote treatment adherence, reduce performance anxiety, and integrate treatments into a sexual relationship. (Moderate Recommendation; Evidence Level: Grade C)
7. Clinicians should counsel men with ED who have comorbidities known to negatively affect erectile function that lifestyle modifications, including changes in diet and increased physical activity, improve overall health and may improve erectile function. (Moderate Recommendation; Evidence Level: Grade C)
8. Men with ED should be informed regarding the treatment option of an FDA-approved oral phosphodiesterase type 5 inhibitor (PDE5i), including discussion of benefits and risks/burdens, unless contraindicated. (Strong Recommendation; Evidence Level: Grade B)
9. When men are prescribed an oral PDE5i for the treatment of ED, instructions should be provided to maximize benefit/efficacy. (Strong Recommendation; Evidence Level: Grade C)
10. For men who are prescribed PDE5i, the dose should be titrated to provide optimal efficacy. (Strong Recommendation; Evidence Level: Grade B)
11. Men who desire preservation of erectile function after treatment for prostate cancer by radical prostatectomy (RP) or radiotherapy (RT) should be informed that early use of PDE5i post-treatment may not improve spontaneous, unassisted erectile function. (Moderate Recommendation; Evidence Level: Grade C)
12. Men with ED and testosterone deficiency (TD) who are considering ED treatment with a PDE5i should be informed that PDE5i may be more effective if combined with testosterone therapy. (Moderate Recommendation; Evidence Level: Grade C)
13. Men with ED should be informed regarding the treatment option of a vacuum erection device (VED), including discussion of benefits and risks/burdens. (Moderate Recommendation; Evidence Level: Grade C)
14. Men with ED should be informed regarding the treatment option of intraurethral (IU) alprostadil, including discussion of benefits and risks/burdens. (Conditional Recommendation; Evidence Level: Grade C)
15. For men with ED who are considering the use of IU alprostadil, an in-office test should be performed. (Clinical Principle)
16. Men with ED should be informed regarding the treatment option of intracavernosal injections (ICI), including discussion of benefits and risks/burdens. (Moderate Recommendation; Evidence Level: Grade C)
17. For men with ED who are considering ICI therapy, an in-office injection test should be performed. (Clinical Principle)
18. Men with ED should be informed regarding the treatment option of penile prosthesis implantation, including discussion of benefits and risks/burdens. (Strong Recommendation; Evidence Level: Grade C)
19. Men with ED who have decided on penile implantation surgery should be counseled regarding post-operative expectations. (Clinical Principle)
20. Penile prosthetic surgery should not be performed in the presence of systemic, cutaneous, or urinary tract infection. (Clinical Principle)
21. For young men with ED and focal pelvic/penile arterial occlusion and without documented generalized vascular disease or veno-occlusive dysfunction, penile arterial reconstruction may be considered. (Conditional Recommendation; Evidence Level: Grade C)
22. For men with ED, penile venous surgery is not recommended. (Moderate Recommendation; Evidence Level: Grade C)
23. For men with ED, low-intensity extracorporeal shock wave therapy (ESWT) should be considered investigational. (Conditional Recommendation; Evidence Level: Grade C)
24. For men with ED, intracavernosal stem cell therapy should be considered investigational: (Conditional Recommendation; Evidence Level: Grade C)
25. For men with ED, platelet-rich plasma (PRP) therapy should be considered experimental. (Expert Opinion)
This guideline's purpose is to provide direction to clinicians and to men who have ED. The guideline focuses on how to recognize ED, how to conduct a valid diagnostic process, and how to approach treatment with the goals of restoring sexual function and enhancing the man and his partner's quality of life (QoL) while minimizing adverse events (AEs) and diagnosis- and treatment-associated burden. The strategies and approaches recommended in this document were derived from evidence-based and consensus-based processes. There is a continually expanding literature on ED; the Panel notes that this document constitutes a clinical strategy; it is intended to be interpreted with appreciation for the dynamic, evolving understanding of ED causes and treatments. The most effective approach for a particular man is best determined by that man (in consultation with his partner, when applicable) in collaboration with the clinician and with full consideration of the relevant history, values, and goals for treatment using a shared decision-making (SDM) approach. As our understanding of ED evolves and improves, the strategies presented here will be amended to remain consistent with the highest standards of clinical care.
Systematic review. A systematic review was conducted to identify published articles relevant to the diagnosis and treatment of ED. Literature searches were performed on English-language publications using the Pubmed, Embase, and Cochrane databases from 1/1/1965 to 7/29/2017. Data from studies published after the literature search cut-off will be incorporated into the next version of this guideline. Preclinical studies (e.g., animal models), commentary, and editorials were excluded. Additional exclusion criteria included data not relevant to current practice (e.g., reports on medications not in current clinical use, outcomes for prostheses models that are no longer available), articles focused primarily on surgical technique with minimal or no patient information or outcomes reported, no outcomes reported or outcomes data not extractable, or duplicate report of data presented elsewhere. Review article references were checked to ensure inclusion of all possibly relevant studies. Multiple reports on the same patient group were carefully examined to ensure inclusion of only non-redundant information. The systematic review yielded a total of 999 publications relevant to preparation of the guideline.
Data on study type (e.g., published systematic review/meta-analysis, randomized controlled trial [RCT], controlled clinical trial [CCT], observational study), treatment parameters (e.g., type of treatment, dosing, follow-up), patient characteristics (e.g., age, symptom duration, ED severity), outcomes (e.g., effects on erectile function, QoL), and AEs were extracted.
Quality of Studies and Determination of Evidence Strength. The quality of published systematic reviews was assessed using A Measurement Tool to Assess Systematic Reviews (AMSTAR). 1 Individual studies that were RCTs or CCTs were assessed using the Cochrane Risk of Bias tool. 2 The quality of case-control studies and comparative observational studies was rated using the Newcastle-Ottawa Quality Assessment Scale. 1004 Because there is no widely-agreed upon quality assessment tool for single cohort observational studies, the quality of these studies was not assessed.
The categorization of evidence strength is conceptually distinct from the quality of individual studies. Evidence strength refers to the body of evidence available for a particular question and includes not only individual study quality but consideration of study design; consistency of findings across studies; adequacy of sample sizes; and generalizability of samples, settings, and treatments for the purposes of the guideline. The American Urological Association (AUA) categorizes body of evidence strength as Grade A (well-conducted and highly-generalizable RCTs or exceptionally strong observational studies with consistent findings), Grade B (RCTs with some weaknesses of procedure or generalizability or moderately strong observational studies with consistent findings), or Grade C (RCTs with serious deficiencies of procedure or generalizability or extremely small sample sizes or observational studies that are inconsistent, have small sample sizes, or have other problems that potentially confound interpretation of data). By definition, Grade A evidence is evidence about which the Panel has a high level of certainty, Grade B evidence is evidence about which the Panel has a moderate level of certainty, and Grade C evidence is evidence about which the Panel has a low level of certainty. 3
AUA Nomenclature: Linking Statement Type to Evidence Strength. The AUA nomenclature system explicitly links statement type to body of evidence strength, level of certainty, magnitude of benefit or risk/burdens, and the Panel's judgment regarding the balance between benefits and risks/burdens (See the Table 1 button below). Strong Recommendations are directive statements that an action should (benefits outweigh risks/burdens) or should not (risks/burdens outweigh benefits) be undertaken because net benefit or net harm is substantial. Moderate Recommendations are directive statements that an action should (benefits outweigh risks/burdens) or should not (risks/burdens outweigh benefits) be undertaken because net benefit or net harm is moderate. Conditional Recommendations are non-directive statements used when the evidence indicates that there is no apparent net benefit or harm or when the balance between benefits and risks/burden is unclear. All three statement types may be supported by any body of evidence strength grade. Body of evidence strength Grade A in support of a Strong or Moderate Recommendation indicates that the statement can be applied to most men in most circumstances and that future research is unlikely to change confidence. Body of evidence strength Grade B in support of a Strong or Moderate Recommendation indicates that the statement can be applied to most men in most circumstances but that better evidence could change confidence. Body of evidence strength Grade C in support of a Strong or Moderate Recommendation indicates that the statement can be applied to most men in most circumstances but that better evidence is likely to change confidence. Body of evidence strength Grade C is only rarely used in support of a Strong Recommendation. Conditional Recommendations also can be supported by any body of evidence strength. When body of evidence strength is Grade A, the statement indicates that benefits and risks/burdens appear balanced, the best action depends on the man's circumstances, and future research is unlikely to change confidence. When body of evidence strength Grade B is used, benefits and risks/burdens appear balanced, the best action also depends on individual man's circumstances and better evidence could change confidence. When body of evidence strength Grade C is used, there is uncertainty regarding the balance between benefits and risks/burdens, alternative strategies may be equally reasonable, and better evidence is likely to change confidence.
For some clinical issues there was little or no evidence from which to construct evidence-based statements. Where gaps in the evidence existed, the Panel provides guidance in the form of Clinical Principles or Expert Opinion with consensus achieved using a modified Delphi technique if differences of opinion emerged. 4 A Clinical Principle is a statement about a component of clinical care that is widely agreed upon by urologists or other clinicians for which there may or may not be evidence in the medical literature. Expert Opinion refers to a statement, achieved by consensus of the Panel, that is based on members' clinical training, experience, knowledge, and judgment for which there is no evidence.
Process. The Male Sexual Dysfunction Panel was created in 2013 by the American Urological Association Education and Research, Inc. The Practice Guidelines Committee of the AUA selected the Panel Co-Chairs who in turn appointed the additional panel members with specific expertise in this area. The AUA conducted a thorough peer review process. The draft guideline document was distributed to 35 peer reviewers. The Panel reviewed and discussed all submitted comments and revised the draft as needed. Once finalized, the guideline was submitted for approval to the Practice Guidelines Committee, the Science and Quality Council, and subsequently to the AUA Board of Directors for final approval. Funding of the panel was provided by the AUA; panel members received no remuneration for their work.
| TABLE 1: AUA Nomenclature Linking Statement Type to Level of Certainty, Magnitude of Benefit or Risk/Burden, and Body of Evidence Strength | |||
| Evidence Strength A (High Certainty) | Evidence Strength B (Moderate Certainty) | Evidence Strength C (Low Certainty) | |
| Strong Recommendation (Net benefit or harm substantial) | Benefits > Risks/Burdens (or vice versa) Net benefit (or net harm) is substantial Applies to most patients in most circumstances and future research is unlikely to change confidence | Benefits > Risks/Burdens (or vice versa) Net benefit (or net harm) is substantial Applies to most patients in most circumstances but better evidence could change confidence | Benefits > Risks/Burdens (or vice versa) Net benefit (or net harm) appears substantial Applies to most patients in most circumstances but better evidence is likely to change confidence (rarely used to support a Strong Recommendation) |
| Moderate Recommendation (Net benefit or harm moderate) | Benefits > Risks/Burdens (or vice versa) Net benefit (or net harm) is moderate Applies to most patients in most circumstances and future research is unlikely to change confidence | Benefits > Risks/Burdens (or vice versa) Net benefit (or net harm) is moderate Applies to most patients in most circumstances but better evidence could change confidence | Benefits > Risks/Burdens (or vice versa) Net benefit (or net harm) appears moderate Applies to most patients in most circumstances but better evidence is likely to change confidence |
| Conditional Recommendation (No apparent net benefit or harm) | Benefits = Risks/Burdens Best action depends on individual patient circumstances Future research unlikely to change confidence | Benefits = Risks/Burdens Best action appears to depend on individual patient circumstances Better evidence could change confidence | Balance between Benefits & Risks/Burdens unclear Alternative strategies may be equally reasonable Better evidence likely to change confidence |
| Clinical Principle | A statement about a component of clinical care that is widely agreed upon by urologists or other clinicians for which there may or may not be evidence in the medical literature | ||
| Expert Opinion | A statement, achieved by consensus of the Panel, that is based on members' clinical training, experience, knowledge, and judgment for which there is no evidence | ||
Definition. ED is defined as the inability to attain and/or maintain penile erection sufficient for satisfactory sexual performance. 5 The Panel also endorses the Fourth International Consultation on Sexual Medicine's ED definition as the consistent or recurrent inability to attain and/or maintain penile erection sufficient for sexual satisfaction. 6
Conceptualization of ED. Sexuality is a uniquely complex aspect of humanness. Sexual function depends on intact anatomical, physiological, and behavioral capacities but occurs in the multi-layered context of a man's beliefs and values about sexuality and maleness, his upbringing and sociocultural mores, his relationship with his partner and the quality of that partnership, and the partner's beliefs and values about sexual activity. No other aspect of human functioning touches upon so many components of a man's identity and leverages this degree of complexity.
In this complex human context, the Panel conceptualizes ED as the inability to attain and/or maintain sufficient penile rigidity for sexual satisfaction. The Panel advocates that awareness of this perspective informs every aspect of the process in which clinicians support and guide men and their partners in evaluation, diagnosis, and choice of management.
Ample evidence indicates that ED is a risk marker for the presence of treatable underlying medical conditions that, left untreated, reduce quality and length of life (e.g., undiagnosed diabetes, CVD). 7 , 8 In addition, ED can negatively affect a man's mental health, his relationship, and his general well-being. The presence of ED, therefore, provides an opportunity to potentially address multiple issues that affect a man's general health.
Shared decision-making (SDM). SDM is the cornerstone of patient-centered care, applying the concepts of autonomy and respect for persons to the clinical encounter. 1005 SDM is a process in which information about the best available evidence for diagnostic procedures and treatments is shared by clinicians and patients. Patients are then supported during the decision-making process to express preferences and values that ultimately lead to an informed choice aligned with those preferences and values. 9 SDM rests on the assumption that individual self-determination is desirable and that patient autonomy is best supported by a strong relationship with an informed and committed clinician who respects the patient's competence and capacity to make decisions. 9 To be effective, this process requires commitments by both clinician and patient. The clinician's commitment includes communicating objectively and clearly regarding the patient's condition and the available diagnostic and treatment options, using language and concepts that are understandable to the patient. 10,11 This commitment includes the awareness that health literacy varies widely across patients and that patients at all levels of health literacy may struggle to objectively apply information about benefits and risks/burdens of various management options. 11 This commitment also requires that the clinician be cognizant that social, cultural, religious, educational, and other factors are important and valid determinants of treatment selection. 12,13 The patient's commitment includes the willingness to absorb information, ask questions, and clearly express his and his partner's preferences and values. This process results in a sharing of information and responsibility, allowing a collaborative decision regarding diagnostic and treatment plans. Because of the complexity of sexuality and the impact of a sexual relationship on a man's life, the Panel strongly advocates that a man's partner be invited to participate in this process whenever possible and clinically appropriate.
Treatment of ED. Although the principles underlying the treatment of ED are the same for all men - restoring or enhancing sexual function, improving overall physical health, and optimizing QoL and well-being for a man and his partner - every man who presents with ED is unique. Each man brings to the clinical encounter not only his symptoms, but his degree of distress; his associated health conditions; his partner's concerns and issues of relationship quality; and his sociocultural, educational, and religious context. Determining an appropriate treatment requires that the man, his clinician, and ideally his partner navigate all of these issues in order to arrive at a treatment choice that is aligned with the man and his partner's priorities and values. Men should be informed of all treatment options that are not medically contraindicated and supported in the SDM process to determine the appropriate treatment. Although many men may choose to begin with the least invasive options (i.e., oral medications), the Panel notes that it is valid for men to begin with any type of treatment, regardless of invasiveness or reversibility. Men also may choose to forego treatment. In each scenario, the clinician's role is to ensure that the man and his partner have full understanding of the benefits and risks/burdens of the various management strategies. All men, regardless of the decision to treat ED, should be strongly advised to address any underlying medical issues that may contribute to the ED and that constitute independent risk factors for poor health, reduced QoL, and decreased survival.
Epidemiology. Up to 30 million men in the United States and 150 million men worldwide are estimated to be affected by ED. 14,15 Independent risk factors for ED and CVD are well recognized and include age, smoking, diabetes mellitus, hypertension, dyslipidemia, depression, obesity, and a sedentary lifestyle. 16-19 Compelling evidence exists that the most common underlying mechanism of ED is vascular and that CVD and ED share etiologies as well as pathophysiology. 20-22 The degree of ED strongly correlates with severity of CVD, and recent studies suggest that ED may be considered a sentinel marker in men with occult CVD. 23, 24 Symptoms of ED may precede a cardiovascular event by up to five years. 25, 26 Further, when ED is present in younger men, it predicts a marked increase (up to 50 fold) in the risk of future cardiac events, suggesting that young men with ED in particular would benefit from CVD risk factor screening and intervention. 27 The increased number of men with CVD risk factors is paralleled by the worldwide increase in the prevalence of ED. 15, 28-30
Hypertension . Hypertension is a highly prevalent condition, affecting 29.1% of U.S. adults between 2011 and 2012. 31 It is frequently associated with ED and often contributes to its etiology (i.e., hypertension-related arterial stenotic lesions). It is present in 38% to 42% of men with ED, and approximately 35% of men with hypertension have some degree of ED. 32-35
Dyslipidemia . Data from the National Health and Nutrition Examination Survey (2003-2006) indicate that approximately 53% of U.S. adults have lipid abnormalities. 36 Up to 42.4% of men with ED also have hyperlipidemia. 33 Elevated levels of total cholesterol and low-density lipoprotein cholesterol are significantly correlated with moderate to severe ED. 33 Men with poor to very poor erectile function had twice the odds of an elevated total cholesterol/high-density lipoprotein cholesterol ratio compared with men with good and very good erectile function. 37
Diabetes mellitus . ED is one of the most common complications of diabetes mellitus. Depending on the severity and duration of diabetes, the prevalence of ED ranges from 20% to 85%. 32,38,39 With a projected increase in the number of patients with diabetes to 29 million by 2050, a corresponding increase in those with ED is also expected. Approximately 20% of men with ED also had diabetes. 33 The Massachusetts Male Aging Study reported a 28% age-adjusted prevalence of ED in men with diabetes compared with 10% in men without diabetes (a 3-fold increased risk). 19 Prevalence of ED is higher in men with diabetes who are older than 50 years, nearly double that in age-matched men without diabetes (45.8% versus 24.1%). In addition, an increase in the relative risk of ED was associated with increased duration of diabetes. 37 ED is known to occur at an earlier age in men with diabetes than in those without it. 38 In some cases, ED may be a manifestation of previously undiagnosed diabetes mellitus, which highlights the importance of screening men with ED for diabetes-related risk factors.
Other non-cardiovascular comorbidities . Other comorbidities or risk factors commonly associated with ED include depression, smoking, premature ejaculation (PE), lower urinary tract symptoms (LUTS) secondary to benign prostatic hyperplasia (BPH), and other causes of voiding dysfunction, such as overactive bladder. In 3 surveys of men with ED, depression was reported by 11% and PE by approximately 30% to 60% of respondents. 33 Several studies have documented a strong association between LUTS and ED. LUTS/BPH, reported in up to 72% of men with ED, are independent risk factors for each other and share both non-cardiovascular (e.g., age, mental disorders) and cardiovascular (e.g., obesity, hypertension, diabetes mellitus) risk factors. 40-43 These relationships underscore the importance of assessing ED in men who present with these common conditions.
The Diagnostic Approach. Insufficient literature was identified to constitute an evidence base for diagnosis of ED in clinical practice. This section, therefore, is based primarily on Clinical Principles or Expert Opinions. This section is intended to provide clinicians and men who present to them with a framework for determining whether a diagnosis of ED is appropriate; it is not intended to replace the judgment and experience of the individual clinician faced with a particular man.
1. Men presenting with symptoms of ED should undergo a thorough medical, sexual, and psychosocial history; a physical examination; and selective laboratory testing. (Clinical Principle)
The sexual response cycle is conceptualized as a sequential series of psychophysiological states that usually occur in an orderly progression; these phases were characterized by Masters and Johnson as desire, arousal, orgasm, and resolution. ED can be conceptualized as an impairment in the arousal phase of sexual response; however, impairments in arousal are likely to have secondary effects on a man's sexual interest and ability to achieve orgasm. 44,45 In addition, men may have an inadequate understanding of the sexual response cycle and may confuse changes in sexual desire, orgasm/refractory period, ejaculatory function (i.e., premature or rapid ejaculation), and conditions such as Peyronie's Disease (PD) with ED. Information regarding reduced or absent libido is important to elicit given that successful ED treatment will not address this issue, and it may continue to generate frustration and anxiety for the man and his partner. For these reasons, thoughtful, detailed, and compassionate inquiry regarding sexual concerns is necessary. Given that many men are uncomfortable broaching the topic of sexual concerns with a physician, it is critical that the physician initiate the inquiry. 46, 47
When the man's presenting concern is ED, a comprehensive evaluation and targeted physical exam should be performed. Detailed assessment of sexual concerns may be difficult in a clinical setting when the presenting complaint is not ED; however, basic inquiry into sexual health should be a standard of care in any encounter in which conditions are discussed or interventions contemplated that may influence a man's sexual life.
Medical, sexual, and psychosocial history . The etiology of ED is often multifactorial. General medical history factors to consider when a man presents with ED are age, comorbid medical and psychological conditions, prior surgeries, medications, family history of vascular disease, and substance use. Common risk factors for ED include vascular disease, tobacco use, neurologic disease, endocrinopathies, medication-related side effects, and psychosocial issues. Vascular issues are particularly important because in some cases they can be improved with lifestyle interventions, such as dietary changes, weight loss, and increased physical activity (see Guideline Statement 7).
Key questions regarding ED include identifying the onset of symptoms, symptom severity, degree of bother, specification of whether the problem involves attaining and/or maintaining an erection, situational factors (e.g., occurring only in specific contexts, only when with a partner, only with specific partners), the presence of nocturnal and/or morning erections, the presence of masturbatory erections, and prior use of erectogenic therapy. 47 The presence of nocturnal and/or morning erections suggests (but does not confirm) a psychogenic component to ED symptoms that would benefit from further investigation. Additional important information includes whether symptoms have been stable or are progressive; worsening symptoms may suggest the presence of progressive underlying comorbidities, particularly cardiovascular comorbidities, that need to be definitively addressed. Categorizing ED severity involves integrating findings from the history and physical, responses to questionnaire content, and any additional diagnostic tests undertaken.
It is important to distinguish ED from PE or early ejaculation, defined as ejaculation before or shortly after penile penetration 48 leading to subsequent loss of erection due to the resolution phase, and from the refractory period, an interval after ejaculation/orgasm in which the penis will not become erect and which tends to increase in duration as a man ages. Information about changes in libido, orgasm, and penile morphology (e.g., the possible presence of PD) also is needed. The timing of specific symptoms should be ascertained in relation to the onset of ED as these symptoms may be primary causes of ED or secondary effects of the ED condition.
The man's sexual partner(s) plays a key role in determining the appropriateness and efficacy of any intervention. 12,47 The ideal clinical situation is one in which the assessments and treatment discussions include the partner. If the man has a partner, then the partner's views on ED and treatment should be assessed, when possible. Additional details, such as the partner's gender, the duration of the relationship, ongoing or unresolved interpersonal/relationship issues, the partner's views on sexuality, and the partner's personal health/sexual issues, are useful to support a man in the evaluation of ED and to select an appropriate management strategy.
Physical exam . Vital signs including pulse and resting blood pressure should be assessed. Obesity is a key indicator of ED risk. 49 Consideration should be given to the assessment of waist circumference. 50 BMI is an alternative but has less specificity for central adiposity, which is a more robust indicator of underlying CVD. The general physical examination should include assessment for signs of TD (e.g., gynecomastia, under-developed facial/pubic/axillary hair). Genital examination should include assessment of penile skin lesions and placement/configuration of the urethral meatus. If the man is considering penile prosthesis implantation or surgical intervention, then documentation of flaccid stretched penile length (a proxy for erect length) can be useful information to guide expectations for outcomes.
Examination of the penis for occult deformities or plaque lesions should occur with the penis held stretched and palpated from the pubic bone to the coronal sulcus. 51 The presence/absence of a palpable plaque should not be taken as definitive evidence for clinically relevant penile deformity such as PD. If PD is suspected, then additional diagnostic procedures should be undertaken (i.e., an in-office ICI test; see AUA Peyronie's Disease guideline). General consistency of the penile tissue can be assessed. Scrotal examination may include general assessment of the scrotal skin and palpation of the testicles to assess for size, consistency, and location. Digital rectal examination (DRE) is not required for evaluation of ED; however, BPH is a common comorbid condition in men with ED and may merit evaluation and treatment. Because BPH/LUTS are commonly comorbid and detected at the same time as ED, appropriate evaluation and therapy for these conditions should be considered. For a detailed discussion of BPH/LUTS, please see the AUA Guideline on this topic. DRE may also permit assessment of the bulbocavernous reflex, which provides information on neural integrity of the pelvis. Absence of the bulbocavernous reflex is not in itself diagnostic, however, as this reflex is absent in up to 30% of normal patients. 51 DRE should be considered for men with TD who may proceed with testosterone therapy.
Selected laboratory tests . With the possible exception of serum testosterone, glucose/hemoglobin A1c, and in some cases serum lipids, no routine serum study is likely to alter ED management. However, serum studies are an important component of evaluation because they may provide information on the etiology of ED and reveal the presence of additional conditions that require treatment. Basic studies appropriate in some men that may be ordered by the treating clinician if recent laboratory results are not available include serum BUN/Cr, fasting lipids, fasting glucose or hemoglobin A1c, and morning testosterone (see Guideline Statement 4). Thyroid function studies (i.e. thyroid-stimulating hormone, free T4) and PSA may be appropriate for some men with ED. If elevated serum PSA is detected during evaluation for ED, then appropriate counseling should occur; please see the AUA guideline on the early detection of prostate cancer for further information. 1006
The importance of psychological factors . Psychological factors (i.e., depression, anxiety, relationship conflict) and psychosexual issues may be primary or secondary contributors to ED. 52,53 Men may not appreciate that depression, anxiety, stress, and relationship conflicts can interfere with the physiological processes necessary for erectile function. Thoughtful discussion of these issues with men and their partners is a key component of patient education and can promote acceptance of incorporating a mental health/sexuality expert into the treatment plan. Involvement of a mental health expert with knowledge and experience to address issues of sexuality with men and their partners can benefit most patients (see Guideline Statement 6) and should be strongly considered when unresolved issues appear to be affecting the sexual relationship. 54-56 In situations in which sudden or severe ED is likely to develop (e.g., men considering definitive therapy for pelvic cancers) or in cases with complex psychosocial issues (e.g., history of sexual trauma, long-term/lifelong sexual dysfunction), early inclusion of psychosexual expertise on the treatment team is critical to development of an effective and feasible treatment plan.
2. For the man with ED, validated questionnaires are recommended to assess the severity of ED, to measure treatment effectiveness, and to guide future management. (Expert Opinion)
Validated questionnaires quantify ED severity and the consequences of ED (e.g., bother, sexual satisfaction, relationship impact). These instruments, or incorporation of their content as part of history and follow-up interviews, are useful to measure treatment effectiveness and to adjust management plans based on outcomes over time. They can be used to quantify unassisted erectile function compared to erectile function with treatment or across treatments (e.g., at a different medication doses). Questionnaires also can provide an opportunity to initiate a conversation about ED when sexual concerns are not the presenting issue. Note that questionnaires will not generate a valid score for the man who is not sexually active. In some settings, a short form validated questionnaire may be most appropriate; examples include the Erection Hardness Score (EHS) 57 and the Sexual Health Inventory for Men (SHIM). 58 The EHS is a single-item instrument that asks men to rate erection hardness on a scale that ranges from 0 (penis does not enlarge) to 4 (penis is completely hard and fully rigid). The SHIM is comprised of five questions scored from 1 to 5; total scores of 22-25 are interpreted as no ED, 17-21 as mild ED, 12-16 as mild-to-moderate ED, 8-11 as moderate ED, and 5-7 as severe ED.
For specialty practices when the presenting issue is ED, a more detailed instrument such as the full form of the International Index of Erectile Function (IIEF) may be more useful. 59,60 Multi-component surveys (e.g. the IIEF) permit a brief but nuanced assessment of sexual function in men. The IIEF consists of 15 questions that quantify 5 domains (sexual desire, erectile function, intercourse satisfaction, ejaculatory/orgasmic function, overall sexual satisfaction). The erectile function (EF) domain quantifies ED severity on a scale of 5-30; scores of 26-30 are consistent with normal erectile function, 18-25 consistent with mild ED, 11-17 consistent with moderate ED, and ≤10 consistent with severe ED. 60 Note that the SHIM is sometimes referred to as the IIEF-5 because it uses five of the six questions that comprise the IIEF-EF subscale, but the interpretation of scoring ranges is different. Clinicians should be aware that clinically significant degrees of erectile function improvement depend on initial symptom severity, with greater improvements necessary for satisfactory results in men with more severe symptoms at baseline. 61 The Male Sexual Health Questionnaire also provides a more in-depth assessment of sexual function. 62 This instrument has 25 questions that constitute subscales for Erection, Ejaculation, and Satisfaction. A four-question version of the Ejaculation subscale also is available to measure ejaculatory dysfunction. 63
3. Men should be counseled that ED is a risk marker for underlying cardiovascular disease (CVD) and other health conditions that may warrant evaluation and treatment. (Clinical Principle)
Risk markers are attributes that predict increased probability of a disease state but are not part of the causal pathway. ED is a risk marker for systemic CVD. 25,26,64 The relationship between ED and clinical CVD was originally posited based on a shared clinical risk factor model (including hypertension, smoking, and diabetes) and the presumed overlap in pathophysiological mechanisms including inflammation, endothelial dysfunction, and atherosclerosis. 65 In the early 2000s, longitudinal studies on CVD and ED suggested a two-way relationship such that patients with CVD are more likely to have ED and patients with ED are more likely to develop future CVD, even when adjusted for shared risk factors. 28,66-68 The Princeton Consensus Conference, an inter-specialty meeting centered on preserving cardiac function and optimizing sexual health, has identified ED as a substantial independent risk marker for CVD. 69 Data from the Prostate Cancer Prevention Trial indicated that the presence of ED was as strong a predictor of future cardiac events as cigarette smoking or a family history of myocardial infarction. 66 Most recently, the QRISK group incorporated ED as an independent risk factor into their updated 10-year cardiovascular risk model, with the presence of ED conferring a 25% increased risk for the average middle-aged man. 70
The diagnosis of ED provides a pivotal opportunity to discuss and address cardiovascular risk. The clinician should communicate this increased risk to the man with ED, to his partner, and to other relevant clinicians (e.g., the primary care provider) so that appropriate referrals and interventions can be discussed and implemented. The diagnosis of ED, and the associated interference with sexual life, may motivate re-evaluation of lifestyle choices and create the motivation for behavioral changes that ultimately may reduce future vascular risks and improve erectile function. 16
Because vascular disease and ED are frequently comorbid, consideration must be given to the possible cardiac risks of sexual activity. Sexual activity has been associated with increased risk for cardiac events, although the absolute risk is small, particularly in men who regularly engage in other physical activities. 71 If there is uncertainty regarding a man's exercise tolerance and fitness for sexual activity, then he should be referred for in-depth evaluation of cardiac reserve by a cardiologist.
The Princeton III criteria provide guidance regarding when further cardiac evaluation is warranted prior to treating ED by designating patients as low-, intermediate-, or high-risk. 69 Low-risk patients may be treated for ED without additional cardiovascular evaluation. Low-risk patients are men without cardiac disease who are able to exercise with no to minimal cardiac symptoms. Low-risk patients also include men with diagnosed cardiac disease who have undergone successful revascularization procedures (e.g., coronary artery stenting, coronary artery bypass graft), men with controlled asymptomatic hypertension, men with low grade heart failure (i.e., New York Heart Association Class I and II heart failure), and men with mild cardiac valve disease. All other men with cardiovascular conditions require a cardiology consultation and additional cardiac evaluation.
4. For men with ED, morning serum total testosterone levels should be measured. (Moderate Recommendation; Evidence Level: Grade C)
Circulating testosterone levels vary substantially among healthy men and are influenced by episodic and diurnal fluctuations, day to day and seasonal variations, the presence of acute and chronic illness, and by medications. 73 These factors may account for the intra-individual variability of approximately 10% on samples drawn from the same person at the same time of day one to three days apart or three months apart. 73 Diurnal variations also are substantial. Late afternoon levels can be approximately 20% lower than morning values in young men, but the difference may be as high as 50% with a much smaller difference in older men. 74,75 Ideally, therefore, samples should be obtained in the morning. 75
Body of evidence strength . Most studies that documented the range of testosterone values in men were observational and many did not focus on testosterone values as a primary outcome.
5. For some men with ED, specialized testing and evaluation may be necessary to guide treatment. (Expert Opinion)
For some men with ED, generally those who present with complex histories, specialized testing and evaluation may be necessary. Situations that may require more detailed evaluation include men with ED who are 1) young, 2) have a strong family history of cardiac illness, 3) have a history of pelvic trauma, 4) have failed prior ED therapies, 5) have a strong likelihood of primary psychogenic etiology, 6) have concomitant PD, and 7) have had lifelong ED.
Specialized testing should only occur if findings will affect management. Testing should be undertaken by an experienced examiner who is familiar with interpretation of results. To minimize burden, it should be established a priori how a given result will be interpreted and used (e.g. to influence management selection, to determine need for specialist referral).
Nocturnal Penile Tumescence and Rigidity testing . Nocturnal penile tumescence testing involves placement of two strain gauges on the penile shaft to measure radial rigidity during sleep. The device is used over several nights' sleep to quantify the number, rigidity, and duration of nocturnal erections. 78,79 The test has been used historically to differentiate psychogenic from organic etiologies for ED, with the presumption that men with psychogenic ED would have preservation of nocturnal penile erections. However, the test is prone to false negatives and may be less useful in men with impaired sleep schedules.
In office testing . ICI testing assesses veno-occlusive function of penis. In ICI testing, an erectogenic agent (e.g., prostaglandin E1, papaverine, and/or phentolamine) is injected into the corpora cavernosa of the penis. 80,81 Erectile response is assessed 5-10 minutes post injection and typically after sexual stimulation (e.g. masturbation, exposure to audiovisual sexual stimulation). For some men, the sympathetic tone and anxiety involved with in-office penile injection may override the injection agent's activity, leading to a false positive diagnosis of ED. 82,83 Repeat dosing is recommended in such cases. 84 In addition to providing information on penile vascular status, in office erectile function testing may be useful to assess for penile deformities such as PD (see AUA Guideline: Peyronie's Disease).
Penile duplex ultrasound (DUS) may be combined with ICI to produce a more detailed and quantitative assessment of penile vascular response, including arterial sufficiency. 80 DUS also permits observation of plaques and/or fibrosis of the tunica and corporal bodies. DUS is a nuanced procedure and should be performed and interpreted only by those urologists with extensive experience and training in the technique.
DUS is currently the gold-standard in penile vascular evaluation as it is minimally invasive and provides robust information about both cavernous arterial inflow and the veno-occlusive capacity of the penis. 80 These data may be useful for the following:
Key parameters derived from DUS include peak systolic velocity ([PSV], cavernosal artery blood flow rate at start of systole) and end diastolic velocity ([EDV], cavernosal artery blood flow rate at the end of diastole). The velocities are measured in cm/s. Some authorities recommend assessment of PSV and EDV prior to ICI and after ICI with sexual stimulation. However, flaccid PSV is a poor predictor of post-ICI PSV. 85
Different cut-points have been applied for PSV and EDV to diagnose arterial insufficiency and veno-occlusive dysfunction. Generally, a PSV 5 cm/s is consistent with veno-occlusive dysfunction. Resistive Index (defined as PSV-EDV/PSV) is an adjunctive assessment of veno-occlusive dysfunction preferred by some experts. Resistive Index values >0.80 have been cited as indicative of normal veno-occlusive function. 86 Interestingly, men with a very low PSV ( 35 cm/s. 87
Biothesiometry is a non-specific term for testing intended to assess for peripheral neuropathies. Biothesiometry has been applied to the penis, most commonly by applying a device that administers vibrations of controlled and consistent intensity. This device is applied at various penile locations (typically glans but possibly other sites), and the minimal amount of vibration intensity detectable by the patient is quantified. This threshold may then be compared to vibration sensitivity on other parts of the body (e.g., fingertips). Lower thresholds for detection imply greater sensitivity and intact peripheral nerves. Vibration to assess mechanoreceptors is the modality most commonly utilized for biothesiometry; however, light touch and nociceptive nerve fibers may also be tested using the application of sharp versus dull and/or cool versus warm stimuli. Biothesiometry may be informative, but there are few data to suggest that it leads to substantive changes in management in most cases.
Invasive testing . Cavernosometry quantifies intracorporal pressure after ICI and is useful primarily for establishing a diagnosis of veno-occlusive dysfunction. Typically, cavernosometry is performed in conjunction with cavernosography (intracorporal installation of a radio-opaque dye), permitting detailed localization of any area(s) of leak. The procedure is performed by cannulation of the corpora by two butterfly needles; pressure measurements are obtained through one cannula whereas the other is used for infusion of an erectogenic agent followed by continuous infusion of injectable saline with or without radio-opaque dye to maintain a rigid erection.
Selective Internal Pudendal Angiography (SIPA) is a means to precisely elucidate the arterial inflow of the penis and is performed after ICI as corporal blood flows are too low in the flaccid state to permit interpretable information. SIPA involves cannulation of the internal pudendal artery and infusion of a radio-opaque dye to delineate penile arterial anatomy. 88 SIPA is indicated in the rare circumstance of a young patient with arterial insufficiency (confirmed by DUS and most frequently the result of trauma) who may be a candidate for a penile revascularization procedure.
Miscellaneous testing . A variety of alternative modalities have been used to assess ED, including pudendal somatosensory evoked potentials, cavernous electromyograph, bulbocavernous reflex time, and sympathetic skin response). The clinical utility of these tests is unclear at this time; they are not recommended for use outside of a research setting. 89
Treatment Framework. The Panel advocates the use of a treatment framework that is not predicated on men progressing through ED treatments in order of invasiveness or reversibility (See Appendix A: Erectile Dysfunction Algorithm in the menu on the left). Although many men may choose to begin with the least invasive options (i.e., oral medications), any type of treatment as an initial treatment is a valid choice. For each treatment, the clinician's role is to ensure that the man and his partner have full understanding of the benefits and risks/burdens associated with that choice.
6. For men being treated for ED, referral to a mental health professional should be considered to promote treatment adherence, reduce performance anxiety, and integrate treatments into a sexual relationship. (Moderate Recommendation; Evidence Level: Grade C)
The Panel conceptualizes ED as the inability to attain and/or maintain sufficient penile rigidity for sexual satisfaction that occurs in the complex psychosocial context that includes a man's background and beliefs about sexuality, his partner, and that partner's values relevant to sexuality. Psychosocial factors inform and influence every aspect of sexual functioning. The Panel notes that placebo effects reliably occur in trials of ED treatments (i.e., PDE5i) in which men meet diagnostic criteria for organically-caused ED; these effects suggest that even men with organically-driven ED are likely to have unmet needs for psychosocial and relationship support during ED treatment.
Psychotherapy and psychosexual counseling focus on helping patients and their partners improve communication about sexual concerns, reduce anxiety related to entering a sexual situation and during a sexual situation, and discuss strategies for integrating ED treatments into their sexual relationship. Many men avoid using ED treatments or discontinue using effective ED treatments because of beliefs about loss of masculinity and distress related to possible failure in a sexual situation. For men with predominantly psychogenic ED, providers should offer a referral to psychotherapy as either an alternative to medical treatment or as an adjunct to medical treatment. Psychogenic ED is generally driven by a man's anxiety related to the ability to achieve an erection. Medical treatments can be effective in these situations, but the addition of psychotherapy or psychosexual counseling may help men to use the medications more effectively and ultimately transition off medical ED therapies.
A diverse group of studies indicates that support and guidance from mental health professionals for the man with ED and his partner can increase the likelihood of treatment success. In trials that evaluated outcomes for medical therapies with and without psychotherapy, outcomes generally were better in the combined treatment groups. For example, Banner and Anderson (2007) randomized 53 men with psychogenic ED to use sildenafil only or sildenafil in combination with couples cognitive-behavioral therapy. After 4 weeks, more men in the combined condition met criteria for success on the IIEF-EF subscale (48%) and the Overall Satisfaction subscale (65.5%) compared to the sildenafil only condition (29% and 37.5%, respectively). 90 When cognitive-behavioral therapy was added to the sildenafil only group, rates of success became comparable to the original combined treatment group. Melnick et al. (2012) randomized 30 men with psychogenic ED to sildenafil only, group psychotherapy only, or a combined condition. 91 Three questions from the Erectile Dysfunction Inventory of Treatment Satisfaction (EDITS) questionnaire were used to assess treatment satisfaction, confidence in engaging in sexual activity, and naturalness in engaging in sexual activity. At both the end of treatment and three months post-treatment, men in the group therapy only group and the combined treatment group had significantly higher scores on all three questions compared to the sildenafil only group. Titta et al. (2006) randomized 57 men to ICI alprostadil for ED post-non-nerve-sparing RP or cystectomy or ICI in combination with sexual counseling. 92 The counseling intervention involved education about successful ICI use and short-term sexual therapy. At 18 months post-surgery, IIEF-EF scores were statistically significantly higher in the combined treatment group (26.5) compared to the ICI only group (24.3) (note that the magnitude of difference is not clinically significant). Scores on the other IIEF subscales also were significantly higher in the combined treatment group compared to the ICI only group. More patients in the combined group were able to transition to sildenafil 100 mg successfully (27.5%) compared to the ICI only group (17.8%). The dropout rate in the combined group was 0% compared to 28.6% in the ICI only group.
In trials that compared combined medical and psychotherapy to psychotherapy only, the combined groups generally also exhibited greater improvement than the single treatment modalities. For example, Wylie et al. (2003) randomized 45 couples to psychotherapy alone or psychotherapy in combination with a vacuum device. 93 A greater proportion of couples reported improvement in the combined treatment group (84%) compared to the psychotherapy only group (60%).
In addition, randomized trials generally reported significant improvement in outcomes with psychological therapies compared to usual care or wait-list control groups. Three trials of internet-based cognitive-behavioral and modified Masters and Johnson therapy all demonstrated improvements in sexual functioning compared to control groups. 94-96 Meta-analysis of six trials that compared group therapy to a wait-list control or to no therapy revealed a significant reduction in ED persistence associated with group therapy (RR = 0.40; 95% CI 0.17-0.98; p<0.05). 97 Three trials reported ED persistence rates at six months of follow-up; the treatment effect continued to be significant (RR = 0.43; 95% CI 0.26-0.72; p<0.05). For men post-prostatectomy, trials reported an increased likelihood of using erectile aids compared to no therapy when men were supported by peer counseling or nurse counseling 98 and better erectile function with lower treatment dropout rates when psychotherapy was added to a medical ED protocol. 99
Body of evidence strength . The strongest available evidence includes one high-quality systematic review and meta-analysis and a small group of randomized studies. The issues complicating interpretation of this literature are the varied therapeutic approaches used, the diverse outcome measures employed, and the range of patient types evaluated. Sample sizes in many trials were small. Trials ranged in quality from low- to high-risk of bias with most trials in the low- to moderate-risk range because of lack of information about randomization and allocation. Overall, although there is consistent evidence that psychological interventions are effective, there is a lack of a sufficient body of evidence of good quality for a particular type of psychological intervention in a particular type of patient group.
7. Clinicians should counsel men with ED who have comorbidities known to negatively affect erectile function that lifestyle modifications, including changes in diet and increased physical activity, improve overall health and may improve erectile function. (Moderate Recommendation; Evidence Level: Grade C)
The presence of ED indicates the likely presence of other conditions, particularly cardiovascular risk factors. A diverse literature that focused on lifestyle interventions, primarily diet and/or exercise interventions, in men with various comorbidities that often are present in the man with ED indicate that these interventions may have small positive effects on erectile function and broader, positive effects on overall health. The man's presentation for evaluation of ED creates an opportunity for the clinician to emphasize to him and his partner the importance of a healthy lifestyle to general health and QoL, but also to support optimal erectile function and increase the probability that ED treatments will be effective.
Men with metabolic conditions . Esposito et al. (2004) randomized obese men with ED (n = 110) without hypertension, diabetes, or hypercholesterolemia to a weight loss and increased physical activity intervention group or to a general information group. 100 After two years, BMI decreased more and physical activity increased more in the intervention group compared to the general information group. Mean IIEF-5 score improved from 13.9 to 17.0 in the intervention group but remained stable in the general information group (13.5 to 13.6). More men in the intervention group achieved an IIEF-EF score of 22 or greater (n = 17) than in the general information group (n = 3). Esposito et al. (2006) randomized men with metabolic syndrome (n = 65) to a Mediterranean or control diet. 101 ED was not an inclusion criterion. At two years of follow-up, men in the intervention group had improved endothelial function and inflammatory markers (C-reactive protein) compared to the control group. IIEF scores increased more in the intervention group (from 14.4 to 18.1) than in the control group (14.9 to 15.2). More men in the intervention group achieved an IIEF-5 score of 22 or higher (n = 13) compared to the control group (n = 2). Esposito et al. (2009) reported on 209 men with ED or men with significant ED risk factors who underwent an intensive lifestyle change intervention (tailored advice regarding how to reduce body weight, increase physical activity, and improve diet quality). 102 The intervention included sessions with a nutritionist as well as individualized guidance on exercise. Control participants were offered general oral and written information about healthy food choices and increasing physical activity without tailored advice. More men in the intervention group had scores indicating no ED at two years (n = 58) compared to the control group (n = 40). Collins et al. (2013) randomized overweight/obese men (n = 185) to a weight loss resource intervention (SHED-IT Resources), the same intervention plus access to a website with e-feedback, or a wait-list control. 103 At six months of follow-up, the two weight loss groups had lost 4.7 and 3.7 kg, respectively. Analysis of only men with ED at baseline (31.2% of sample) indicated a significant mean 3.3 point increase in the IIEF-5; the wait-list group had a mean decrease of 0.9 points. The authors note that this trial involved no face-to-face contact with participants and no prescribed dietary or exercise regimes. Khoo et al. (2010) randomized obese men with uncomplicated diet or oral hypoglycemic-treated type 2 diabetes (n = 25) or without diabetes (n=19) to a low calorie diet using meal replacements and compared them to a third group of obese non-diabetic men on a control diet. 104 ED was not an inclusion criterion. After eight weeks, IIEF-5 scores increased significantly (from 17.8 to 20.0 in the non-diabetic group and from 8.1 to 10.3 in the diabetic group) for the two intervention groups but not for the control group. Khoo et al. (2013) placed 90 obese men on a low calorie diet and randomized them to perform moderate-intensity exercise (< 150 min/week) or high-intensity exercise (200-300 min/week). 105 At six months follow-up, the men in the high-intensity group had greater increases in the IIEF-5 (from 18.1 to 20.7) compared to the low-intensity group (18.3 to 20.1), but the difference between groups was small (0.8 points). Measures of free testosterone, serum sex hormone-binding globulin, and serum total testosterone also improved in the high-intensity group. Wing et al. (2010) randomized 372 overweight men with type 2 diabetes to a diabetes support and education group or to an intensive lifestyle intervention group that involved individual and group sessions to reduce weight and increase physical activity. 106 These data are from a subset of men who participated in the Look AHEAD trial and completed the IIEF at baseline and at one year of follow-up. At one year, the intensive intervention group had lost more weight and was more fit than the support group. IIEF-EF scores improved more in the intensive intervention group than in the support group, but the magnitude of improvement was small - 17.3 to 18.6 in the intensive group and 18.3 to 18.4 in the support group. In the intensive group, 22% reported an improvement of ED, 70% stayed the same, and 8% reported worsening symptoms. In the support group, 23% reported improvement, 57% stayed the same, and 20% reported worsening symptoms.
Men with cardiovascular conditions . Lamina et al. (2009) randomized 50 hypertensive men with ED to an interval exercise training intervention or a control condition. 107 Men who were obese, had diabetes, smoked, or had other cardiac or renal conditions were excluded. Exercise was performed in three sessions per week for eight weeks. The exercise group had greater improvements in the IIEF-EF (11.5 to 15.1) compared to the control group (8.1 to 8.9), but note that the exercise group's end of treatment score remains in the moderate ED range. Begot et al. (2015) randomized 86 men who had experienced a recent myocardial infarction to a home walking program or a usual care control group; most men (84%) had ED. 108 After one month, 93% of men in the control group had some degree of ED as measured by the IIEF-EF, with 44% having severe ED, 33% having moderate ED, and 16% having mild to moderate ED. In contrast, only 12% of men in the walking group had ED, and all were in the mild category. Kalka et al. (2013, 2015) evaluated 138 men who had been treated invasively for ischemic heart disease and who scored 21 or less on four questions from the IIEF-5. 109,110 Men were randomized into a cardiac rehabilitation group or no rehabilitation control group. Cardiac rehabilitation consisted of interval endurance training three times a week and general fitness and resistance training twice a week. After six months, the mean score on the four IIEF-5 questions was significantly higher in the intervention group (14.4) compared to the control group (12.4). Overall, these data suggest that dietary changes, weight loss, and physical activity increases improve overall health, ameliorate comorbidities associated with ED, and result in small improvements in erectile function overall and may lead to clinically significant improvements in a subset of men. In addition, the Panel notes that given ample evidence that cigarette smokers are at a higher risk of developing ED, men who smoke should be counseled regarding the overall health benefits of smoking cessation. 111
Body of evidence strength . Although most of the available studies are randomized trials, diverse patient populations were evaluated. These include men who are overweight/obese, have metabolic syndrome or type 2 diabetes, or who have various types of cardiovascular conditions. Most trials were not designed with ED as a primary outcome; therefore, not all patients had ED. Lifestyle interventions varied in types of exercise and dietary changes as well as in duration of the interventions. The trials in this group of studies ranged in quality from high- to low-risk of bias with about half in the high-risk category because of lack of information about randomization and allocation. Although it appears to be generally true that improvements in overall health may also improve erectile function, there is a lack of a sufficient body of evidence of good quality for a particular intervention in a particular type of patient group.
8. Men with ED should be informed regarding the treatment option of an FDA-approved oral phosphodiesterase type 5 inhibitor (PDE5i), including discussion of benefits and risks/burdens, unless contraindicated. (Strong Recommendation; Evidence Level: Grade B)
The FDA-approved oral PDE5i available for management of ED in the U.S. include sildenafil, tadalafil, vardenafil, and avanafil. Several other PDE5i have been approved for use in other countries.
The mechanism of action for all commercially available PDE5i is similar. PDE5i inhibit the phosphodiesterase type 5 enzyme from breaking down cyclic guanasine monophosphate (cGMP). This inhibition results in an increase in the concentration of penile cavernosal cGMP that then causes smooth muscle relaxation in the corpus cavernosum vasculature resulting in increased erection hardness and duration in men with ED who have sufficient intact vasculature.
Contraindications . The use of nitrate-containing medications in combination with a PDE5i can cause a precipitous drop in blood pressure; men taking nitrates regularly should not use PDE5i medications. Men who carry sublingual nitroglycerin for angina should be advised not to use this medication within 24 hours of taking a PDE5i, and possibly longer in the case of use of a PDE5i with a long half-life (i.e., tadalafil). Many other medications also potentially can interact with or influence the metabolism of PDE5i, including anti-depressants, anti-fungals, anti-hypertensives, and HIV/AIDS drugs. The clinician who prescribes PDE5i must be conversant with all potential medication contraindications.
For detailed discussion of cardiovascular contraindications to PDE5i use, see guidance from the Princeton III guidelines. 69 In men with mild to moderate hepatic or renal impairment or men with spinal cord injury, PDE5i should be used with caution at least initially at lower doses given the potential for delayed metabolism. In men with severe renal or liver disease, use of PDE5i is generally not recommended.
Efficacy . The PDE5i medications have been extensively studied; nearly a quarter of a million men have been evaluated from the general ED population # and approximately 25,000 men evaluated from various special populations* (e.g., diabetes, BPH/LUTS, post-prostatectomy, post-spinal cord injury). Most studies of men from the general ED population involved on-demand use of medications (approximately 90%). Of the four FDA-approved PDE5i drugs, most men were administered sildenafil or vardenafil with relatively fewer men administered tadalafil and limited data available on avanafil (fewer than 2,000 men).
The general ED population. Data from individual studies and trials, including analyses that pooled data across multiple trials 112-303 and reports of published systematic reviews 304-319 suggest the following major findings:
| TABLE 2: General ED Population: Change in IIEF-EF Scores from Pre-Treatment Baseline to Post-Treatment | ||||
| Treatment | # study arms^ | Minimum | Maximum | Mean |
| Placebo | 62 | -1.60 | 7.10 | +1.78 |
| Sildenafil | 26 | +1.70 | +11.75 | +9.00 |
| Tadalafil | 37 | +1.98 | +12.00 | +7.82 |
| Vardenafil | 26 | +5.30 | +12.90 | +8.80 |
| Avanafil | 5 | +5.50 | +9.40 | +8.10 |
Special populations. Fewer studies focused on special populations, but in general, findings are similar to those reported in the general ED population. 320-499 For example, the available data suggest that the PDE5i have similar efficacy. Note, however, that not all PDE5i have been evaluated in all special populations of men with ED. For men with diabetes, sildenafil, tadalafil, and vardenafil appear equally effective with limited data reported for avanafil. For men with BPH/LUTS and ED, sildenafil and tadalafil appear to have similar efficacy to treat ED. There are no studies of vardenafil or avanafil that focused on men with BPH/LUTS and ED. All studies of men with BPH/LUTS and ED used daily dosing because of the beneficial urinary tract effects of PDE5i. For men with ED post-prostatectomy, efficacy also appears similar across the PDE5i but with limited data for avanafil. For men post-RT for prostate cancer, sildenafil and tadalafil appear to have similar efficacy, but the tadalafil data are limited. No studies evaluated vardenafil or avanafil in men post-RT for prostate cancer. For other special populations (i.e., spinal cord injury, renal transplant) there are insufficient data for different PDE5i to come to a definitive conclusion. Overall, there are insufficient data at different PDE5i doses to evaluate dose-response effects in special populations. The data suggest, however, that men with diabetes and men who are post-prostatectomy have more severe ED at baseline and respond less robustly to PDE5i (for detailed discussion of men who have undergone treatment for prostate cancer, see Guideline Statement 11).
AEs. Most AEs associated with the administration of PDE5i are mild to moderate. The most frequently reported AEs were dyspepsia, headache, flushing, back pain, nasal congestion, myalgia, visual disturbance, and dizziness (data for avanafil are limited; see Appendix B for detailed tables). On average, rates of dyspepsia and dizziness were relatively similar across sildenafil, tadalafil, and vardenafil. There were sufficient dyspepsia data to meta-analyze. When studies were collapsed across medications, the Relative Risk (RR) for dyspepsia in an active treatment arm compared to placebo was 3.21 (95% CI 2.5-4.3; p
Tadalafil was the only medication for which there were substantial on-demand versus daily dosing studies. Generally, daily dosing (which allows men to take a lower dose) was associated with lower rates of frequently-reported AEs - particularly for headaches - compared to on-demand use, which requires a higher dose (see Appendix B). Meta-analysis of headache data indicated that on-demand dosing was associated with a significantly higher risk of headache (RR = 2.65; 95% CI 1.8-3.8; p0.05; I 2 = 0%; risk not significantly different from placebo groups).
Most AEs followed a dose-response pattern such that men in active treatment arms reported statistically significantly higher rates of AEs than did men in placebo arms, and the percentage of men reporting a particular AE increased as dose increased. Within individual studies, however, the differences between dose groups were usually not statistically significantly different.
When means for the general and four special populations (men with diabetes, BPH/LUTS, post-RP, or post-RT) for which there are substantial data were examined, it appears that men post-RP and men post-RT reported substantially higher rates of AEs than did men in the general ED population (see Appendix B). Whether men who have had prostate cancer treatment are more likely to experience AEs or are more likely to report AEs is not clear. Men post-RP reported higher rates of AEs in response to sildenafil than in response to other PDE5i. Men post-RT reported high rates of AEs across PDE5i and in placebo groups. The high rates of AEs reported by men in placebo groups suggest that men post-RT may have heightened sensitivity to body sensations and may have unmet needs for psychosocial support.
Other concerns. Nonarteritic anterior ischemic optic neuropathy (NAION). NAION is a rare visual condition characterized by the sudden onset of loss of vision in one eye. The estimated annual incidence is 2.5 to 11.8 cases per 100,000 men aged 50 years or older with older age, Caucasian ethnicity, small optic discs with low cup-to-disc ratio, and various kinds of vascular conditions appearing to confer greater risk. 500-502 Several studies have suggested that PDE5i use is associated with an increased risk of NAION, 503-505 although the absolute risk is small (3 additional cases per 100,000 men aged 50 years or older. 506 Men in higher-risk groups (e.g., older men, men of Caucasian ethnicity, men with vascular risk factors) should be counseled about this small increased risk, including the fact that the absolute risk of NAION is extremely low with or without the use of PDE5i, and that the association does not imply causation.
Skin cancers. Several investigations have addressed the possible relationship between PDE5i use and increased risk for skin cancers, particularly malignant melanoma. Most papers report a small positive association for at least one indicator of increased risk. 507-511 However, none of the available studies was designed to adequately address the potentially important confounders of increased medical surveillance among PDE5i users that could result in more frequent detection of skin cancers and the possibility that men who take PDE5i have more ultraviolet radiation exposure than do non-PDE5i using men. The issue of ultraviolet radiation exposure via sunlight is suggested by the positive association between PDE5i use and increased risk of basal cell carcinoma and solar keratosis - both conditions related to sun exposure. 511 Further, men with a history of solar keratosis, an indicator of high sun exposure, were more likely to become PDE5i users. 511 Overall, the available findings fail to convincingly satisfy most of Hill's causal criteria (i.e., strength, consistency, specificity, temporality, biological gradient in which higher levels of exposure increase risk, plausibility) for determining whether an epidemiological association constitutes a causal relationship. 512 The Panel interpreted these data to indicate that there is no increased risk of skin cancers reliably associated with PDE5i use.
Prostate cancer recurrence. Several studies have focused on the possible relationship between PDE5i use after prostate cancer treatment and an increased risk of prostate cancer recurrence. The initial report of this relationship indicated that PDE5i use was an independent risk factor for prostate cancer recurrence among men with localized disease who underwent bilateral nerve-sparing RP. 513 However, three subsequent studies that performed a more nuanced analysis of this relationship (i.e., assessed dose-response relationships) did not confirm this finding 514-516 for men post-RP or post-RT. In contrast, all three studies reported that PDE5i use non-significantly reduced prostate cancer recurrence rates. The Panel interpreted these data to indicate that there is no increased risk of prostate cancer recurrence associated with PDE5i use after prostate cancer treatment.
Body of evidence strength . Body of evidence strength for outcomes and AEs associated with PDE5i therapy for the general ED population is Grade B. Most of the data were provided by RCTs that ranged in quality from poor (high risk of bias based on the Cochrane rating system) to high-quality (low risk of bias), with the majority of RCTs rated as moderate-quality (unclear risk of bias). The most frequent reason for a rating of unclear risk of bias was inadequate information regarding randomization and/or blinding. Strengths of this group of studies are the use of randomization, blinding, and placebo control groups to protect internal validity and the extremely large sample sizes. The weaknesses of these studies are in two areas. First, approximately 70% of studies had industry associations in terms of authorship affiliations and/or financial support. Second, most trials (approximately 75%) ended at three months or less of follow-up. Given that PDE5i use is likely to continue for periods much longer than three months, there is an important gap in information regarding the long-term treatment consequences.
Published systematic reviews, meta-analyses, and network analyses also constitute an important source of evidence. The body of published systematic reviews on the effects of PDE5i medications in the general ED population constitutes Grade B evidence. AMSTAR scores, which range from 1 to 11 points and quantify the methodological quality of the systematic review, ranged from 3 to 10 points with one-half of studies scoring 6 or less. The most common deficits were inadequate study selection and data extraction procedures, failure to use quality ratings in interpreting findings, failure to report or address heterogeneity across individual studies, and failure to assess for publication bias. Most published systematic reviews exhibited an additional weakness that can affect validity - collapsing data across general and special populations. In addition, all systematic reviews were handicapped by the fact that many RCTs did not provide sufficient information for data to be included in a meta-analysis.
Body of evidence strength for special populations was Grade C. Limited numbers of randomized studies were available for specific subgroups, limiting conclusions.
9. When men are prescribed an oral PDE5i for the treatment of ED, instructions should be provided to maximize benefit/efficacy. (Strong Recommendation; Evidence Level: Grade C)
Men who are prescribed a PDE5i should be carefully instructed in the appropriate use of the medication. In particular, it should be explained that sexual stimulation is necessary and that more than one trial with the medication may be required to establish efficacy. It should also be explained that the medications differ in onset of action, duration of action, and whether food intake limits efficacy (see the Table 3 below).
| TABLE 3: Characteristics of PDE5i Medications | |||
| PDE5i | Onset of action | Duration of action | Effect of food intake |
| Avanafil | 15-30 min | Up to 6 hours | Not affected |
| Sildenafil | 30-60 min | Up to 12 hours | High-fat meal decreases efficacy |
| Vardenafil | 30-60 min | Up to 10 hours | High-fat meal decreases efficacy |
| Tadalafil | 60-120 min | Up to 36 hours | Not affected |
Studies of men who report non-response to PDE5i indicate that incorrect use (e.g., lack of sexual stimulation, medication taken with a large meal) accounts for 56% to 81% of treatment failures. 517-520 When men were re-educated regarding appropriate medication use, including medication-specific requirements, from 23.6% to 58.5% experienced treatment success.
10. For men who are prescribed PDE5i, the dose should be titrated to provide optimal efficacy. (Strong Recommendation; Evidence Level: Grade B)
When prescribing a PDE5i, the clinician must balance these priorities: the goals of the man and his partner for successful sexual activity, the need to prescribe an effective PDE5i dose, and the need to minimize AEs. It is important for clinicians, men who desire a PDE5i, and partners to communicate regarding how treatment success is defined. The clinician's goal is to work with the man and his partner to find the dose that meets treatment expectations without resulting in unacceptable levels of AEs. Although in the context of fixed-dose clinical trials, dose groups generally did not exhibit statistically significantly different average response levels. For the individual patient, dose titration is a key step to optimize efficacy. This process may require that initial doses are titrated up or down until the optimal dose is identified. To minimize distress, men and partners should be counseled that initial non-response or inadequate response may be readily overcome with a dose increase just as initial unacceptable levels of AEs may be ameliorated with a dose decrease. Given that men with diabetes or post-prostatectomy often present with more severe levels of ED, clinicians may consider initiating therapy at a higher dose.
The clinician should be aware that when PDE5i studies were examined in aggregate, the differences in response rates between dose groups were extremely small, rarely statistically significant, and generally not clinically significant. For example, in studies in which men were administered 50 mg sildenafil, pre-treatment IIEF-EF scores averaged 14.0 and post-treatment scores averaged 22.6; in studies in which men were administered 100 mg sildenafil, pre-treatment IIEF-EF scores averaged 14.4 and post-treatment scores averaged 23.8 (see table in Appendix B). This pattern in which large dose increases (i.e., doubling the dose) resulted in small differences in average response level or rate was consistent across the PDE5i (see discussion under Guideline Statement 8).
In contrast, although reported AE rates vary considerably from study to study, on average AE rates generally increased as dose increased (see discussion under Guideline Statement 8 and data in Appendix B). The goal of dose titration, therefore, is to achieve patient- and partner-defined success while minimizing AEs.
As part of the process of identifying the optimal dose, men may be offered dosing frequency changes or different PDE5i. For example, Kim, Seftel et al. (2013) evaluated 623 men who had suboptimal results with on-demand maximum dose sildenafil, tadalafil, or vardenafil; men were offered daily tadalafil (2.5mg or 5mg) or placebo for 12 weeks. 521 Approximately 40% of men in the tadalafil groups achieved IIEF-EF scores indicating normal erectile function (IIEF-EF ≥26) compared to 12% in the placebo group. Carson, Hatzichristou et al. (2004) reported that when men unresponsive to sildenafil were randomized to flexible dose vardenafil (5mg to 20mg) or placebo, positive responses to the Sexual Encounter Profile (SEP) 2 question doubled (from 30.3% at baseline to 62.3% post-treatment), and positive response to the SEP 3 question quadrupled (from 10.5% at baseline to 46.1% post-treatment). 522 However, these approaches will not benefit men with severe arterial insufficiency. 523-525
For men who appear to have ED primarily or entirely of psychogenic origin, consideration should be given to dose reduction and/or medication weaning once treatment of the psychological issues has occurred and confidence has been restored.
Body of evidence strength . Body of evidence strength for outcomes and AEs associated with dose titration of PDE5i therapy is Grade B. See discussion under Guideline Statement 8.
11. Men who desire preservation of erectile function after treatment for prostate cancer by radical prostatectomy (RP) or radiotherapy (RT) should be informed that early use of PDE5i post-treatment may not improve spontaneous, unassisted erectile function. (Moderate Recommendation; Evidence Level: Grade C)
In the modern era of prostate cancer management, improving functional outcomes, particularly sexual function, has become a priority. Accordingly, "penile rehabilitation" or "erectile function rehabilitation" has emerged as a clinical management practice that focuses on preserving erectile capability that is at risk of decline during treatment of pelvic malignancies such as prostate cancer. 526,527 Although this practice is widely accepted as a general concept, it is variously defined. 528 In strict terms, "penile rehabilitation" comprises strategic approaches that promote natural erectile capability and facilitate resumption of medically unassisted sexual activity after prostate cancer treatment. 529 However, more broadly considered, "penile rehabilitation" encompasses the application of interventions in any form that address the negative effects of cancer treatment on erectile ability as well as related health aspects. 526,527,530 This practice differs conceptually and practically from treating ED that is present post-prostate cancer therapy with oral or other therapies.
Clinically localized treatments such as RP and RT as well as systemic therapies used for advanced disease (e.g., hormonal therapy), result in various degrees of ED. Although erectile function outcomes in these contexts have improved over time, many men will experience clinically significant ED as a consequence of prostate cancer treatment. With respect to RP, for example, the development of cavernous nerve-sparing surgical procedures (i.e., the application of techniques that preserve the peri-prostatic penile nerve supply required for penile erection) has led to improved rates of erectile function recovery, 531,532 but even with use of this technique many men will experience ED. 533-535 A meta-analysis of studies with >12 months follow-up post-RP reported that use of a bilateral nerve-sparing technique was associated with a 60% erectile function recovery rate (95% CI 58.0 - 62.0; 21 studies) compared to a rate of 47% (95% CI 42.0 - 53.0; 12 studies for use of a unilateral nerve-sparing technique. 536 For RT, modifications in the delivery of radiation have resulted in better erection preservation after treatment, 537 but rates of new-onset ED have been reported at 36% and 38% two and three years post-RT, respectively. 538
The natural history of erectile function loss and recovery depends on the type of prostate cancer intervention. The classically observed immediate effects of RP on penile erection are absent responses under all stimulatory conditions. 539,540 When cavernous nerves are spared, a gradual recovery of erectile function is possible, although this recovery may be delayed for several months at a minimum. Commonly, the interval of spontaneous erectile function recovery occurs 12 to 24 months after surgery, although recovery may still be possible as much as 36 months after surgery. 541 RP studies indicate that while improvements in erectile function may occur over time post-operatively, relatively few men recover baseline erectile function, particularly those over age 60 years at the time of surgery. 542 When cavernous nerves are not spared, which may occur when wide excision of locally advanced prostate cancer is necessary or when nerve-sparing attempts are inadequate, the expected effect is an unrecoverable loss of erectile function. 539,540 The natural history of erectile impairment after radiation, in contrast, involves a delayed onset of ED that may occur 24 to 36 months after treatment and may worsen over time thereafter. 543
The pathophysiology of post-prostatectomy ED principally involves "neuropraxia" (i.e., temporary traumatic functional loss of nerve function) that may occur despite "nerve-sparing," or complete nerve function loss that occurs after cavernous transection or removal. In addition to nerve injury, concomitant injury of accessory penile vasculature and secondary structural and functional derangements of the denervated cavernosal tissue may contribute to ED. 527 The pathophysiology of post-radiation therapy ED involves radiation-induced damage of the nerve and vascular supply of the penis. 544
Strategies for penile rehabilitation aim to prevent or reduce the extent of long-term erectile impairment and/or latency of erectile function recovery. The objective of these strategies is to counteract pathophysiologic mechanisms of ED induced by prostate cancer treatments. Theorized objectives at the tissue or cellular level include improving oxygenation of cavernosal tissue, promoting protection of penile vascular and sinusoidal endothelial function, and reducing cavernosal tissue damage associated with the penile denervation effect. 545 The rationale for their use on the molecular level centers on biological principles of sustaining and modulating nerve function (i.e., neuroprotection, neurogenesis, and neuroregeneration) as well as vascular/cavernosal tissue function (i.e., vasculoprotection, vasculogenesis, angiogenesis and anti-fibrosis). 546-548
A variety of treatments have been introduced as penile rehabilitation strategies, with prescribed recommendations for their implementation that consider timing, schedule, and delivery of treatment. It is strongly perceived that rehabilitation should be initiated at the beginning of if not before prostate cancer treatment with its continuation over a prescribed interval thereafter. The rationale for this timing is that an early, preventative scheme is maximally protective of mechanisms of penile erection. 526,527,545 A common strategic approach has been to apply erectogenic treatments used to treat ED according to precise rehabilitative protocols. This approach is based on the theory that induced penile vascular blood flow enhances the functional status and recovery of erectile tissue. 545
PDE5i have been investigated most extensively for the purpose of penile rehabilitation because of their non-invasiveness and ease of administration. Several rigorous randomized, CCTs have been performed in settings of RP and pelvic irradiation. 429,430,438,470,526,527 These trials have not demonstrated that early PDE5i use (i.e., within 45 days of prostate cancer therapy) improves unassisted erectile function. Meta-analysis performed for this guideline of the three trials that compared a PDE5i to placebo among men who had RP yielded a pooled RR of 1.0 (95% CI 0.66 - 1.64; p=0.85; nonsignificant heterogeneity), indicating no difference in rates of restored erectile function between groups. In addition, although most studies reported that PDE5i are effective in assisting erections on-demand during the course of the trial, early administration of PDE5i does not improve later responses to these medications compared to early administration of placebo.
One possible confounding issue is whether the treatment schedule of 12 months or less in these trials was insufficient to provide erectile function rehabilitative benefit; it is possible that a longer term treatment schedule may be necessary to achieve erectile health recovery effects. Use of other types of therapy in a rehabilitative framework, such as VED or ICI, also has been reported. 549 Although some studies reported positive findings, given the small numbers of men treated and the lack of control groups in several studies, more mature data are needed to establish these therapies as proven.
Overall, the Panel interpreted the PDE5i data to indicate that erectile function rehabilitative protocols tested to date remain unproven. Psychosocial support, however, is an important strategy for penile rehabilitation. Given the impact of ED after prostate cancer treatment, particularly its suddenness and severity for many men undergoing RP, it is not surprising that men in this setting commonly experience depression, anxiety, and relationship stress. 550,551 Psychotherapeutic regimens have been prescribed with reported rehabilitative benefits of such treatment. 526 Clinicians should educate men regarding the sexual effects of prostate cancer treatments and set realistic expectations regarding functional recovery, including the possibility that recovery may be more challenging for men who have multiple ED risk factors. Men should be coached and monitored during and after prostate cancer treatments. 552 These efforts, including combining psychosocial support and somatic erectogenic treatments, may motivate men and their partners to maintain intimacy during sexual function recovery. 526,530,551
Body of evidence strength . Most of the available randomized trials were rated as having an unclear risk of bias (moderate quality) because of lack of information regarding randomization and/or blinding. Whether a sufficient duration of treatment has been tested to achieve erectile function restoration is unclear.
12. Men with ED and testosterone deficiency (TD) who are considering ED treatment with a PDE5i should be informed that PDE5i may be more effective if combined with testosterone therapy. (Moderate Recommendation; Evidence Level: Grade C)
Modes of testosterone administration varied across trials and included oral testosterone, testosterone patch, and testosterone gel. Criteria for testosterone deficiency also varied across trials. Primary outcome measures were the IIEF, IIEF-EF, or SHIM. Across trials, men who received combined therapy reported greater erectile function score increases compared to baseline levels and higher erectile function scores at treatment end than did men who received a PDE5i alone or testosterone alone. Meta-analysis performed for this guideline of the four studies that compared combined treatment to a PDE5i alone yielded a weighted mean difference of 2.69 points on the IIEF-EF between treatment groups (95% CI 0.96-4.41; p=0.002; nonsignificant I 2 ). An additional group of observational studies reported that the addition of testosterone to a PDE5i among men in whom the PDE5i alone was ineffective resulted in improved erectile function. 558-562 Although the differences between the combined and monotherapy groups in the randomized studies were not statistically significant in all trials, the Panel interpreted these data to indicate that for symptomatic testosterone deficient men, optimum efficacy of PDE5i medication is most likely to be achieved when testosterone levels are normalized. Similar conclusions were reported by three published systematic reviews. 308,563,564
The Panel notes that it is likely that the restoration of testosterone levels supports maximum efficacy of other ED treatment options; however, there are insufficient data at this time to address other combined treatments.
Men should be advised that testosterone therapy is not an effective mono-therapy for ED. 565-567 If the man's goal is amelioration of ED symptoms, then he should be counseled regarding the need for ED therapies in addition to testosterone therapy. However, testosterone therapy may provide more global health benefits (e.g., improved bone density). For detailed information on possible health benefits of testosterone therapy, AEs associated with testosterone therapy, and recommended monitoring protocols for men prescribed testosterone, see AUA guideline on this topic. 1007
Body of evidence strength . The best evidence consists of five randomized trials and three published systematic reviews. The consensus across reviews is that the available trials are mostly of low quality. Four trials compared combined testosterone + PDE5i treatment with PDE5i only treatment; one trial compared combined treatment to testosterone only treatment. The trials differed in mode of testosterone administration and in PDE5i dosing regimen (i.e., on-demand versus daily). Trial sample sizes were small with three of five trials evaluating samples smaller than 40 men (two trials had sample sizes of 10 per treatment group). Definitions of TD differed across trials and follow-up durations were short, ranging from one to 3.5 mos.
13. Men with ED should be informed regarding the treatment option of a vacuum erection device (VED), including discussion of benefits and risks/burdens. (Moderate Recommendation; Evidence Level: Grade C)
Vacuum devices are associated with high rates of patient and partner satisfaction and are an effective and low-cost treatment option for select men with ED. They are effective in the general ED population as well as in men with diabetes, spinal cord injury, post-prostatectomy, and other conditions. 93,496,568-617 Only VEDs containing a vacuum limiter (a feature that limits the amount of vacuum pressure and reduces the potential for penile injury) should be used, whether purchased over-the-counter or procured via prescription.
Studies on VED satisfaction and efficacy largely pre-date the era of the IIEF, EDITS, and the Self-Esteem And Relationship Questionnaire (SEAR), etc. Clinicians should be aware that many studies were carried out before the availability of PDE5i medications, and some studies suggest that when men have a choice, more men prefer PDE5i 496,618 The most commonly-reported outcome measure was in terms of a responder criterion and/or patient and partner satisfaction rates (see the Table 4 button below). Responders were usually defined as men who obtained an erection sufficient for intercourse with use of the device although some studies defined responders as men who purchased the device after a trial period or who continued to use the device. Rates for patient and partner satisfaction and for successful responses exhibited a wide range but the majority of studies reported high rates. Of the 12 studies that reported patient satisfaction rates, six rates were 80% or higher and eleven studies reported rates of 60% or higher. Of the seven studies that reported partner satisfaction rates, all rates were above 70% except for one. Of the 28 studies that reported a success criterion, 19 reported rates of 75% or higher. Twenty-five studies reported rates of 56% or higher. Lewis, Witherington (1997) performed a survey of approximately 6,000 VED users and reported that 75% remained continuous users, 83.5% reported having sex as frequently as desired, and 70% reported improved relationships (note that this survey was performed before the introduction of PDE5i). 587 One study reported findings in terms of IIEF scores. Khayyamfar, Forootan (2013) reported on 1,530 men at an unspecified follow-up duration. 583 Statistically significant improvements in all the IIEF subscales were reported with vacuum device use. These authors also reported that 92.7% of patients successfully used the device to have intercourse.
In men who are PDE5i non-responders, VED may have a role as a "rescue" device. A study of 69 men who had failed PDE5i therapy and were subsequently treated with a combination of PDE5i and VED for 4 weeks demonstrated a significant increase in the IIEF, SEP 2 and 3, and the Global Patient Assessment Scale indicating that this regimen may be effective in at least a subset of this cohort. 619
Men with diabetes . One randomized design 608 and six observational studies 602-607 evaluated the use of vacuum devices in diabetic men. The randomized design compared men who used a vacuum device with 100 mg sildenafil to men who used only a vacuum device; these patients were non-responders to sildenafil alone, and all had Type 2 diabetes. Two studies evaluated a mixed group of Type I and II diabetes patients. One study compared patients with Type I to patients with Type II diabetes. Three studies did not specify diabetes type. Follow-up durations ranged from 2 months to two years. Sample sizes were small except for Israilov et al. (2005), which began with 162 patients. 605 The duration of diabetes ranged from 5.3 years to 17.5 years.
Sun, Peng (2014) reported that men in the VED + sildenafil group had larger increases in SHIM scores and reported higher rates of "yes" responses to the SEP 2 and 3 than did men in the VED only group. 608 Five observational studies reported outcomes in terms of patient satisfaction (two studies - range 81.2 to 84%), partner satisfaction (two studies - range 72.7 to 80%), and/or a successful responder criterion. These varied from the achievement of an erection sufficient for intercourse to an undefined "positive response." Successful responses ranged from 70.4% to 90%. Pajovic, Dimitrovski (2017) reported outcomes in terms of IIEF scores; at 6 months scores on the IIEF-EF subscale had increased significantly among men with Type I and men with Type II diabetes as had scores on the intercourse satisfaction subscale and the overall satisfaction subscale. 606
Men post-prostatectomy . Three randomized designs and one observational study evaluated the use of VED in men who were post-prostatectomy. Kohler, Pedro (2007) randomized men who had undergone unilateral or bilateral nerve-sparing RP to early (one month postoperatively) or late (six months postoperatively) use of a VED. 620 Engel (2011) compared men post-bilateral nerve-sparing prostatectomy who used tadalafil to men who used tadalafil + vacuum device. 609 Raina, Agarwal (2006) compared no treatment to use of a vacuum device in a group of men who had bilateral, unilateral or non-nerve-sparing prostatectomy. 611 Nason, McNamara (2016) also evaluated a mixed group of men post-RP. 610 The three randomized studies began treatment approximately one-month post-RP with the goal of optimizing unassisted erectile function. The observational study began treatment 8.7 months post-RP with the goal of treating ED. Sample sizes were extremely small across studies.
Kohler, Pedro (2007) reported that unassisted (i.e., without use of the device) IIEF-EF scores at 3 and 6 months were significantly higher in the early intervention group compared to the late group (11.5 and 12.4 versus. 1.8 and 3.0) and that stretched penile length was preserved in the early intervention group but reduced by mean 2 cm in the late intervention group. 620 Engel (2011) reported that men in the tadalafil + vacuum group had higher unassisted SHIM scores (18.9) compared to men in the tadalafil only group (11.1). 609 Similarly, 92% of men in the combined treatment group achieved an erection sufficient for penetration compared to 57% of men in the tadalafil only group. Raina, Agarwal (2006) reported higher SHIM scores for men who used a vacuum device (16.0) compared to men who had no treatment post-RP (11.2). 611 The observational study reported that 81.8% of men achieved an erection sufficient for intercourse. 610
Clinicians should counsel men with ED prior to beginning VED treatment about the potential occurrence of AEs. Most AEs were minor and resolved without intervention. The most commonly-reported AEs were transient penile petechiae or bruising (16 studies: mean 17.7%; range 0 to 50%), discomfort or pain (17 studies: mean 18.2%; range 0 to 64%), difficulty with ejaculation (9 studies: mean 21.6%; range 3.4 to 40%), and difficulty with the device (10 studies: mean 19.8%; range 0 to 66.6%). Some men also noted loss of sensitivity (7 studies: mean 14.5%; range 3.2 to 45%). Men who are receiving anti-coagulant therapy and/or who have bleeding disorders or have a history of priapism should use VEDs with caution. 621
Body of evidence strength . More than 90% of study arms were contributed by observational designs. Inclusion criteria varied and limited information was reported regarding patient characteristics such as the severity of ED or the presence of comorbidities. Most studies pre-date the era of validated questionnaires (e.g., the IIEF, SEAR, EDITS) and the era of PDE5i. AE reporting was variable with most studies not indicating the severity of AEs. Study dropout rates complicate interpretation because only successful patients continued to use the devices. Sample sizes were small. Although approximately 11,000 men participated in vacuum device studies, two post-marketing surveys accounted for more than half of these patients (approximately 7,000 men). 587, 601
14. Men with ED should be informed regarding the treatment option of intraurethral (IU) alprostadil, including discussion of benefits and risks/burdens. (Conditional Recommendation; Evidence Level: Grade C)
IU medication involves the insertion of a delivery catheter into the meatus and depositing an alprostadil pellet in the urethra to induce an erection sufficient for intercourse. Alprostadil is prostaglandin E1. IU alprostadil is a treatment option for men for whom PDE5i are contraindicated, for men or partners who prefer to avoid oral medication, and/or for men or partners who prefer not to use the needles required for ICI medications.
In the general ED population, two RCTs, 622,623 one randomized design that compared IU alprostadil to ICI alprostadil, 624 one open-label crossover that compared IU alprostadil to ICI alprostadil, 625 and a group of observational studies, 626-635 evaluated this medication.
Importantly, most studies proceeded with chronic treatment only in men who had erections firm enough for intercourse in response to in-office testing. The success rates among men who used the medication chronically, therefore, are relevant to responsive intra-office testing - not men with ED in general. In-office positive testing rates across studies exhibited a large range, from 20% to 65.9%. Clinicians and men with ED should be aware that a large proportion of men who have a positive in-office test will not be successful in the home environment.
Successful intercourse rates (variously defined across studies) with IU alprostadil ranged from 29.5% to 78.1%. The largest study to assess the efficacy of IU alprostadil reported that 995 of 1,511 (65.8%) men had positive responses in the office; only men with positive in-office responses were then randomized to the IU alprostadil or placebo groups. Of the 461 men assigned to the alprostadil condition, only 299 (64.9%) achieved at least one episode of intercourse at home, 622 indicating that a positive in-office test does not guarantee efficacy in the home environment. In addition, only 73% of doses (2,634 of 3,593) were successful in facilitating intercourse, orgasm, or a 10-minute erection sufficient for intercourse.
Overall, in the two placebo-controlled studies, success rates were statistically significantly higher in the IU alprostadil conditions compared to the placebo condition. In the two studies that compared IU alprostadil to ICI alprostadil, success rates were significantly higher in the ICI group. In the four studies that reported outcomes using the IIEF subscales or the SHIM, 627,629,634,636 scores after treatment were significantly higher than pre-treatment baseline scores but generally were not indicative of normal erectile function. Men should be counseled that that IU approach is generally less effective than the ICI approach. 624,636
AEs were frequently reported but minor and short-term. The most common AEs were genital pain (ranging from 6.5 to 34.7%), minor urethral trauma (ranging from 1 to 5.1%), urethral pain or burning (0 to 29%), and dizziness (0 to 7.0%). Episodes of hypotension or syncope were rare. One study reported that 1% of men experienced an episode of prolonged or painful erection. 623 There were no reports of priapism.
Body of evidence strength . The randomized studies were of moderate quality (unclear risk of bias), but follow-up durations were short at three months. Measures of success across studies were defined in ways that are not clearly comparable. These include the occurrence of intercourse at least once during the study, the occurrence of intercourse at least twice during the first two months of the study, 75% of erections adequate for intercourse, erection sufficient for intercourse without need for an additional soft rubber band, improved erection quality or increased intercourse frequency or improved SHIM score, percentage of men who continued to use the medication at study end, and percentage of successful medication uses. Approximately 69% of studies reported an authorship or funding association with the pharmaceutical industry.
15. For men with ED who are considering the use of IU alprostadil, an in-office test should be performed. (Clinical Principle)
IU alprostadil should not be prescribed until a man has undergone instruction in the method, an initial dose-titration in the office, and detailed counseling regarding possible AEs and actions to take in response to potentially serious AEs.
IU alprostadil is available in doses of 100 μg, 250 μg, 500 μg, and 1,000 μg. The clinician should select a dose for in-office testing that is expected to produce an erection sufficient for intercourse. The higher the dose, the more likely the man will experience an AE; therefore, the lowest dose expected to be effective should be used. Instructions for application include urinating before use because residual urine in the urethra aids in dissolution and dispersal of the medicine along the urethra. The penis is then pulled straight and held pointing up. The applicator stem is placed approximately 3 cm into the urethra and the button is depressed. The applicator is moved slightly to separate the pellet from the applicator tip and the applicator is removed. The penis is kept upright and rolled between the hands to aid in dissolution and dispersal of medication. The man is advised to walk or stand for approximately 10 min to aid in blood flow.
Although episodes of priapism were not reported in IU alprostadil trials, the man should be thoroughly educated about priapism and instructed on safe responses and maneuvers in a prolonged erection situation. Commonly-used strategies (but for which no evidence was retrieved) include attempting ejaculation and, if this effort is unsuccessful, oral pseudephedrine followed by the application of an ice pack to the penis for 30 minutes to an hour. If a painful, non-bendable erection persists after these strategies, then the man should proceed to the emergency room within two to four hours of medication administration.
16. Men with ED should be informed regarding the treatment option of intracavernosal injections (ICI), including discussion of benefits and risks/burdens. (Moderate Recommendation; Evidence Level: Grade C)
ICI medications are administered by injecting a substance into the corpus cavernosa of the penis to produce an erection. The four substances commonly used in clinical practice are alprostadil, papaverine, phentolamine, and atropine. Only alprostadil is FDA-approved in the U.S. for ICI injection, and it is the only medication typically used as a single agent. Combinations of medications also are used (i.e., papaverine + phentolamine, alprostadil + papaverine + phentolamine; alprostadil + papaverine + phentolamine + atropine). The choice of medication or medication combination is a collaborative decision among the man, partner, and physician and depends on what agent or agents produce an adequate response without unacceptable AEs.
Men who have contraindications to the use of PDE5i, men who prefer not to take an oral medication, or men who find that PDE5i are inadequate or ineffective may choose the ICI approach to treating ED. PDE5i are ineffective in about 40% of men. 216 In addition, a significant proportion of men initially responsive to PDE5i eventually will become non-responsive as ED progresses and will require a different ED treatment approach. Further, a subset of men who find PDE5i effective prefer the ICI alternative. 637
ICI medications have been reported to be effective in diverse groups of men, including men from the general ED population as well as among men with other conditions such as diabetes, cardiovascular risk factors, men who are post-prostatectomy, and men with spinal cord injuries. 624,636,638-771 The most commonly used outcome measure in ICI studies was the percentage of men who reported achieving an erection sufficient for successful intercourse. These percentages ranged from 53.7% to 100% without marked differences across medications or medication combinations. The second most commonly used outcome measure was the percent of men who reported being satisfied with the treatment. These percentages ranged from 46.3% to 98.8% with the lowest satisfaction rates associated with papaverine use (mean 53.4%).
Note that these two outcome measures reflect different qualities of treatment. Achieving an erection sufficient for intercourse focuses on whether the medication produced the desired physiological outcome. Asking whether a man is satisfied with the treatment is a broader, more complex question and may include whether AEs occurred, the partner's views about the mode of administration, the man's comfort with injections, etc.
Relative efficacy of specific medications or medication combinations is difficult to determine because men who are administered single medications or combined medications also differ in presenting ED severity. A stepped care approach can be used to maximize the proportion of men who are treated successfully with ICI. For example, Baniel, Israilov (2000) reported on 625 men who entered a progressive treatment program that used four ICI protocols: (1) papaverine + phentolamine; (2) alprostadil; (3) papaverine + phentolamine + alprostadil; (4) atropine + papaverine + phentolamine + alprostadil. 646 A positive response was defined as erection sufficient for penetration. Positive responses were achieved by 66.4% of the 625 men administered protocol 1. The remaining 210 men were administered protocol 2 (n=75), protocol 3 (n=135) and protocol 4 (n=37). These groups achieved success rates of 36%, 72.6%, and 59.5%, respectively. Only 15 of the 625 men failed to respond to any of the protocols. At three years of follow-up in 610 men, 43.7% had achieved successful intercourse (n=65 without an injection and n=202 with injections).
AEs . Although rates of successful intercourse are similar across medications and medication combinations, AE profiles in the extracted data differed. Men should be thoroughly counseled regarding the potential differential risk profiles of the various ICI substances (see Appendix B). The most serious AE associated with ICI medications is priapism. Most study authors defined priapism as an erection that required intervention in order to resolve; prolonged or painful erections were generally defined as resolving without intervention. The lowest rates of priapism (mean 1.8%) were reported in studies using alprostadil as a single medication (but note that studies of alprostadil reported a mean rate of 6.3% for prolonged or painful erections). The Panel notes that identifying the appropriate dose of medication and thoroughly instructing the man in dose titration is critical to minimize the risk of priapism regardless of the medication or medication combination selected.
Pain is a common consequence of ICI injections; in published studies men reported pain described as pain with injection, penile pain, and genital pain. The literature suggests that pain rates are highest when papaverine (high rates of pain with injection) or alprostadil (high rates of pain with erection) are used as single agents, and when papaverine is used in combination with phentolamine (see table in Appendix B), but there are relatively few studies that used other medication combinations and reported on the incidence of pain.
Penile fibrosis or plaque and penile deformities have been reported with use of ICI. There is considerable range across studies in these reports without any single medication or medication combination clearly associated with higher risk. In addition, the percentage of men who reported these AEs did not increase with follow-up duration. In the absence of reliable predictors for these issues, the Panel suggests that any pre-existing fibrosis or plaque or deformity be documented before initiating ICI and that men be monitored regularly for progression of these conditions or for the onset of a new condition.
Body of evidence strength . More than 90% of study arms were contributed by observational designs - the weakest design in terms of controlling for confounders. Limited information was reported regarding patient characteristics such as the severity of ED or the presence of comorbidities. Most studies pre-date the era of validated questionnaires (e.g., the IIEF, SEAR, EDITS) and rely on patient reports of outcomes. Adverse event reporting was variable with most studies not indicating the severity of AEs. Study dropout rates also complicate interpretation because only successful patients continued to use the medications.
17. For men with ED who are considering ICI therapy, an in-office injection test should be performed. (Clinical Principle)
Men considering ICI therapy should first have an in-office injection test to determine the appropriate dose and medication(s) to produce sufficient duration of response and to minimize AEs. The in-office experience also is important to help the man achieve confidence with the technique and to facilitate adherence. It may take several visits to determine the correct drug(s) and titrate the dose. Men should be informed that although injectable non-prostaglandin agents have been used to successfully manage ED for decades, none are formally FDA-approved for this indication. This visit also should include educating men and their partners regarding how to titrate the dose, the advisability of alternating sites with each dose, and how to proceed if a serious AE occurs (i.e., priapism). The man should be thoroughly educated about priapism and instructed in actions to take in a prolonged erection situation. It is recommended all education be documented. Commonly-used strategies (but for which no evidence was retrieved) include attempting ejaculation and, if this effort is unsuccessful, then oral pseudephedrine followed by the application of an ice pack to the penis for 30 minutes to an hour. If a painful, non-bendable erection persists after these strategies, then the man should proceed to the emergency room within 2-4 hours of medication administration.
The Panel notes that there can be significant cost differences across medications based on how they are obtained (e.g., brand name medications versus compounded medications). Optimizing the medication choice for a particular man ideally includes discussion of costs.
ICI Combination Medications . ICI combination therapy was developed to improve efficacy as a result of the synergistic effects of the drugs and to reduce side effects as a result of using lower dosages of each agent. One complexity encountered with the use of combination medications is the need for the pharmacy to compound these agents because there are no combination ICI drugs currently approved by the FDA. In addition, some substances (e.g., alprostadil) may have a limited shelf-life. 772
No standardized mixture is approved by the FDA; these combinations must be compounded by the pharmacy based on physician instructions. Concentrations of each component vary widely in the literature, but ratios of 12-30 mg papaverine: 10-20 μg alprostadil:1 mg phentolamine are common. A standard dose regimen includes a mixture of 30 mg papaverine + 10 μg alprostadil + 1 mg phentolamine per 1 mL with a starting dose of 0.1-0.5 mL.
18. Men with ED should be informed regarding the treatment option of penile prosthesis implantation, including discussion of benefits and risks/burdens. (Strong Recommendation; Evidence Level: Grade C)
Another choice for the man with ED is the surgical implantation of a penile prosthesis. Prosthesis implantation has been performed successfully in men from the general ED population as well as men from a variety of special populations. 773-885 Men and their partners should be thoroughly counseled regarding the benefits and potential risks of this treatment to ensure appropriate choice of device, realistic post-operative expectations, and high levels of satisfaction. 886 The man and his partner should understand that several devices are currently available, including malleable (non-inflatable) models as well as two- or three-piece inflatable prostheses. The benefits of prostheses include the ability to generate an erection sufficient for intercourse on-demand, for as long as is desired, and as frequently as is desired. The potential risks and burdens of prosthesis surgery include the risks inherent in the surgical procedure, possible changes in the appearance of the penis, and the potential for device malfunction or failure. Men should understand that this treatment choice is best conceptualized as irreversible; although prostheses can be removed, it is unlikely that a man's penis will be reliably responsive to other ED therapies after prosthesis explant.
The most commonly-used outcome measure among this group of studies was the percentage of men who reported being satisfied with prosthesis surgery. The mean satisfaction rate across studies of implanted inflatable models was 86.2%. The mean satisfaction rate across studies of implanted malleable models was somewhat lower at 75.1%. When studies were broken down by prosthesis model groups, satisfaction rates for inflatable models ranged from 85.6% to 88.3% and from 66.1% to 88.7% for malleable models (see table in Appendix B; some studies did not specify prosthesis models).
A smaller number of studies reported partner satisfaction rates (See Appendix B in the menu on the left). Rates were generally high for inflatable models (AMS 700 series - 83.3% and studies that used other, multiple, or unspecified inflatable models - 88.2%), and the AMS Spectra malleable model (89.5%).
A subgroup of studies reported outcomes using the IIEF or the EDITS. These studies also suggest high levels of satisfaction and functionality. 773-776,779,783,793,826,828,850,870,873,874,882,884 In this group of studies, post-operative IIEF-EF scores ranged from 21.8 to 28.8, post-operative SHIM scores ranged from 20.0 to 22.5, and post-operative patient EDITS scores ranged from 57.0 to 90.5, with nine of 11 studies reporting values >75.
AEs . Men and their partners should be counseled regarding AEs. Commonly-reported AEs in the early peri- and post-operative period include penile edema or hematoma (23 studies: range 0.2% to 13.4%; mean 3.4%), corpus injury (11 studies: range 0.06% to 6.2%; mean 2.3%), urethral injury (9 studies: range 0% to 3.1%; mean 1.2%), acute urinary retention (9 studies: range 0% to 4.2%; mean 2.0%), and crura injury (7 studies: range 0.02% to 4.0%; mean 1.5%). These AEs were rarely serious and generally resolved with supportive care or minimal intervention (i.e., short-term use of an indwelling catheter to manage acute urinary retention). Pain in the early post-operative period is not well-documented in the literature, but in the Panel's experience, most men will experience some degree of pain after surgery with complete resolution within one to three months.
Infection. Infection is a serious AE that typically occurs within the first three months after surgery and usually requires removal of the prosthesis. Although no randomized studies have compared outcomes between prosthesis models with and without infection-inhibiting coatings, observational studies indicate that coated models have greatly reduced infection rates with most series reporting rates of 1-2% when these models are implanted. For example, Serefoglu et al. (2012) used patient information forms to compare the Coloplast Titan model with the hydrophilic coating (n=29,360) to the same model without the hydrophilic coating (n=7,031). 867 The infection rate was significantly lower (1.4%) with the hydrophilic coating compared to no coating (4.6%). Similarly, Carson et al. (2011) used 39,005 patient information forms to assess revision for infection in antibiotic-impregnated inflatable devices compared to non-inflatable devices at up to 7.7 years of follow-up. 787 Revision rates for antibiotic-impregnated devices were significantly lower at 1.1% (n = 35,737) than those for non-impregnated devices at 2.5% (n = 3,268). In a retrospective chart review, Droggin, Shabsigh (2005) compared AMS 700 series devices with Inhibizone (n=58) to devices without Inhibizone (n=94). 799 Infection rates for the Inhibizone devices were significantly less (0%) compared to the non-Inhibizone devices (3.2%). Eid et al. (2012) examined infection rates among men implanted with the Coloplast Titan model or the AMS 700 series (results not separated by model), which were without any infection-inhibiting coating (n=132) or had an infection-inhibiting coating (n=704). 801 Infection rates were 5.3% in the non-coated models and 1.99% in the coated models. In this study, a third group of men had coated models implanted, and the surgeons also used a "no-touch" technique. The "no touch" technique involves discarding all surgical instruments and changing all surgical gloves after an incision is made in the penoscrotal raphe and the dissection is carried down through the subcutaneous tissue and dartos to the level of Buck's fascia. Among 1,511 men who were implanted with an infection retardant coated device and who had the "no touch" technique, the infection rate was 0.46%. Antibiotic coatings also appear to reduce infection rates when used to replace a prosthesis. Nehra et al. (2012) reported that at up to 6.6 years of follow-up, secondary revisions as a result of infection were significantly less likely to occur among patients with antibiotic-impregnated replacement implants (2.5%; n = 9,300) compared to non-impregnated implants (3.7%; n = 1.764). 887
This pattern also was evident among diabetic men. At up to 7 years of follow-up in a group of 6,695 diabetic men, significantly fewer patients experienced revision as a result of infection with use of an antibiotic-impregnated prosthesis (1.5%; n = 6,071) compared to men who received non-impregnated models (4.2%; n = 624). 888
Christodoulidou and Pearce (2016) conducted a systematic review to assess whether diabetic men were more vulnerable to infection with prosthesis implant compared to non-diabetic men. 889 The authors noted that most case series reporting higher infection rates among diabetic men date from the 1970s to 1990s and reported rates of 5.5 to 20%; these studies were small. Studies published in the 1990s reported on larger case series and noted lower infection rates, but rates were as high as 10.6%. In 2001, with the use of antibiotic coated implants, infection rates dropped further, with most studies reporting rates of 2% or less. In particular, the sample of 1,511 men described in Eid et al. (2012) who received coated implants using the "no-touch" technique and had an infection rate of 0.46% included 41% diabetic men. 801 The authors conclude that there is no relevant current evidence that diabetic men are at higher risk of prosthesis infection than men from the general ED population.
In select cases, an infected prosthesis can be removed, the location of the device washed out using an antibiotic salvage procedure and a new device immediately placed. This approach should be restricted to men without evidence of sepsis or severe local infection. More typically, the infected device is removed, the infection is addressed with antibiotics, and the tissues are allowed to heal (for six weeks to six months). Once healing has occurred, a new prosthesis may be implanted. However, delayed replacement of a prosthesis after initial removal is a complex operation and it is possible that device placement will not be feasible because of scarring. In addition, in this scenario other problems such as penile shortening, change in penile shape, and loss of sensation are more likely to occur.
Erosions. Erosion or cylinder extrusion occurs when the tissues at the tip of the penis are weakened, allowing the prosthetic cylinder to migrate into the head of the penis and requiring surgical repair and reposition. Erosion rates were on average lower for inflatable models (20 studies: range 0% to 6.5%; mean 2.5%) than for malleable models (7 studies: range 0% to 17.5%; mean 4.1%).
Mechanical failure. Mechanical failure is most common with inflatable models and most likely to occur when a component (usually the connecting tubing) ruptures, resulting in a fluid leak. Numerous refinements in prosthesis design and materials over time have resulted in decreased failure rates. Recent reports suggest that 90% to 95% of men will have a functioning prosthesis 10 years post-surgery. For example, Mirheydar et al. (2016) reported on 5- and 10-year cumulative reoperation rates for 7,666 men with first implant between 1995 and 2010 using the California Office of Statewide Health Planning and Development database. 845 Most men had an inflatable device implanted (88.4%). The total reoperation rate was 11% (904 men), but only 54% of these revision surgeries (in 488 men) were undertaken because of mechanical failure. Of the mechanical failure revisions, more than half involved pump malfunction followed by malfunction of the cylinders and the reservoir. Enemchukwu et al. (2013) examined patient information forms submitted for the AMS 700CX and LGX/Ultrex models; 55,013 devices were implanted between 1997 and 2008, including 39,443 CX devices and 14,470 Ultrex/LGX devices. 802 Devices with and without parylene coating were compared. For CX models, revision rates for mechanical failure at 8.4 years of follow-up were 11.8% for the non-parylene coated device and 6.2% for the parylene coated device. For the Ultrex/LGX models, at 7 years of follow-up mechanical failure revision rates were 7.7% for the non-coated device and 5.5% for the coated device.
Changes in penis appearance. Several studies have reported that some men perceive that the penis is shorter post-implant when the prosthesis is inflated compared to a full erection before the surgery. Few studies, however, have actually measured penile length before and after surgery. Deveci, Martin (2007) measured stretched penile length of 56 men undergoing a first inflatable implant before the surgery and six months post-operatively. 890 Although 72% of men reported that penile length was decreased, the pre- to post-surgery measurements were statistically indistinguishable for the entire group (baseline - 5.2 inches; six months - 5.1 inches) as well as for the group that reported subjective shortening (baseline - 5.1 inches; six months - 5.2 inches). Wang, Howard (2009) compared erect penile length (EPL) induced by ICI injection pre-surgery to erect length after inflatable prosthesis implant. 891 Before surgery, EPL in response to ICI was mean 13.2 cm (5.2 inches); at 6 months and 1 year post-surgery, EPL was 12.5 cm (4.9 inches). These studies suggest that when objective measures are used, small length decreases may be documented.
It is imperative that clinicians discuss and document expectations of post-prosthesis penile length with men and their partners prior to surgery to ensure appropriately calibrated post-operative expectations. In addition, common reasons why men might perceive penile shortening should be discussed. These include possibly inaccurate memories regarding penile dimensions when ED has been present long-term, loss of tissue elasticity over time from the long-term absence of a full erection, weight gain in the pubic area that partially obscures the penis, and the fact that inflation of the prosthesis will not result in glans engorgement, which may make the penis appear to be shorter. Men who have a history of conditions causing tunical scarring, corporal fibrosis, or loss of cavernous smooth muscle should be informed that the prosthesis is unlikely to restore penile dimensions to those present before these conditions occurred.
Several pre-surgical, intraoperative, and post-operative strategies have been examined to maximize penile length and girth after implant. These include the use of pre-surgical penile traction to maximize pre-operative length, 892 the use of pre-surgical VED therapy to facilitate easier corporeal dilatation 893 or to allow longer cylinder placement at the time of surgery. 894 Use of a VED pre-surgically to soften corporeal fibrosis in men with a history of ischemic priapism or infection to facilitate successful implant of a device also has been reported. 895 There also is some evidence that surgeons who are frequent implanters use longer cylinders (median 2 cm longer) compared to surgeons who are less frequent implanters. 896 Several intra-operative techniques also have been examined, including ventral phalloplasty and suspensory ligament release. 781,844,897 . Post-operatively, IU alprostadil and PDE5i medications have been used to improve glans temperature, sensation, and enlargement. 898-900 Successful penile dimension enhancement by using aggressive cylinder sizing and daily cylinder inflation post-operatively with maximal cylinder inflation for one-two hours during post-operative months 6 to 24 also has been reported. 817,901 The Panel notes that currently there are insufficient data on specific approaches and techniques to constitute a reliable evidence base from which to provide clinical guidance regarding these approaches.
Body of evidence strength . The available data were contributed by observational designs and the majority of studies were retrospective. Limited information was reported regarding patient characteristics such as the severity of ED or the presence of comorbidities. Most studies did not use validated questionnaires (e.g., the IIEF, SEAR, EDITS) and rely on patient report or medical chart review. AE reporting was variable and sparse with many studies not addressing AEs, and of the studies that did address AEs, most did not indicate the severity of AEs. Many studies reported large numbers of patients lost to follow-up, creating uncertainty regarding whether additional longer-term AEs (i.e., mechanical failure) may have occurred.