Endocrine evaluation consisting of a morning serum testosterone is generally indicated. Measurement of serum prolactin may be indicated. A low testosterone level merits repeat measurement together with assessment of luteinizing hormone (LH), follicle-stimulating hormone (FSH), and prolactin levels. Other tests may be helpful in excluding unrecognized systemic disease and include a complete blood count, urinalysis, creatinine, lipid profile, fasting blood sugar, and thyroid function studies.

Although not indicated for routine use, nocturnal penile tumescence (NPT) testing may be useful in the patient who reports a complete absence of erections (exclusive of nocturnal "sleep" erections) or when a primary psychogenic etiology is suspected. Such testing should be performed by those with expertise and knowledge of its interpretation, pitfalls, and usefulness. Various methods and devices are available for the evaluation of nocturnal penile tumescence, but their clinical usefulness is restricted by limitations of diagnostic accuracy and availability of normative data. Further study regarding standardization of NPT testing and its general applicability is indicated.

After the history, physical examination, and laboratory testing, a clinical impression can be obtained of a primarily psychogenic, organic, or mixed etiology for erectile dysfunction. Patients with primary or associated psychogenic factors may be offered further psychologic evaluation, and patients with endocrine abnormalities may be referred to an endocrinologist to evaluate the possibility of a pituitary lesion or hypogonadism. Unless previously diagnosed, suspicion of neurologic deficit may be further assessed by complete neurologic evaluation. No further diagnostic tests appear necessary for those patients who favor noninvasive treatment (e.g., vacuum constrictive devices, or pharmacologic injection therapy). Patients who do not respond satisfactorily to these noninvasive treatments may be candidates for penile implant surgery or further diagnostic testing for possible additional invasive therapies.

A rigid or nearly rigid erectile response to intracavernous injection of pharmacologic test doses of a vasodilating agent (see below) indicates adequate arterial and veno-occlusive function. This suggests that the patient may be a suitable candidate for a trial of penile injection therapy. Genital stimulation may be of use in increasing the erectile response in this setting. This diagnostic technique also may be used to differentiate a vascular from a primarily neuropathic or psychogenic etiology. Patients who have an inadequate response to intracavernous pharmacologic injection may be candidates for further vascular testing. It should be recognized, however, that failure to respond adequately may not indicate vascular insufficiency but can be caused by patient anxiety or discomfort. The number of patients who may benefit from more extensive vascular testing is small, but includes young men with a history of significant perineal or pelvic trauma, who may have anatomic arterial blockage (either alone or with neurologic deficit) to account for erectile dysfunction.

Studies to further define vasculogenic disorders include pharmacologic duplex grey scale/color ultrasonography, pharmacologic dynamic infusion cavernosometry/ cavernosography, and pharmacologic pelvic/penile angiography. Cavernosometry, duplex ultrasonography, and angiography performed either alone or in conjunction with intracavernous pharmacologic injection of vasodilator agents rely on complete arterial and cavernosal smooth muscle relaxation to evaluate arterial and veno-occlusive function. The clinical effectiveness of these invasive studies is severely limited by several factors, including the lack of normative data, operator dependence, variable interpretation of results, and poor predictability of therapeutic outcomes of arterial and venous surgery. At the present time these studies might best be done in referral centers with specific expertise and interest in investigation of the vascular aspects of erectile dysfunction. Further clinical research is necessary to standardize methodology and interpretation, to obtain control data on normals (as stratified according to age), and to define what constitutes normality in order to assess the value of these tests in their diagnostic accuracy and in their ability to predict treatment outcome in men with erectile dysfunction.

WHAT ARE THE EFFICACIES AND RISKS OF BEHAVIORAL, PHARMACOLOGICAL, SURGICAL, AND OTHER TREATMENTS FOR IMPOTENCE? WHAT SEQUENCES AND/OR COMBINATION OF THESE INTERVENTIONS ARE APPROPRIATE? WHAT MANAGEMENT TECHNIQUES ARE APPROPRIATE WHEN TREATMENT IS NOT EFFECTIVE OR INDICATED?

General Considerations

Because of the difficulty in defining the clinical entity of erectile dysfunction, there have been a variety of entry criteria for patients in therapeutic trials. Similarly, the ability to assess efficacy of therapeutic interventions is impaired by the lack of clear and quantifiable criteria of erectile dysfunction. General considerations for treatment follow:

• Psychotherapy and/or behavioral therapy may be useful for some patients with erectile dysfunction without obvious organic cause, and for their partners. These may also be used as an adjunct to other therapies directed at the treatment of organic erectile dysfunction. Outcome data from such therapy, however, have not been well-documented or quantified, and additional studies along these lines are indicated.
• Efficacy of therapy may be best achieved by inclusion of both partners in treatment plans.
• Treatment should be individualized to the patient's desires and expectations.
• Even though there are several effective treatments currently available, long-term efficacy is in general relatively low. Moreover, there is a high rate of voluntary cessation of treatment for all currently popular forms of therapy for erectile dysfunction. Better understanding of the reasons for each of these phenomena is needed.