Even though there is no general agreement on the definition of low testosterone levels, in the general population, 24 % of men have morning serum total testosterone levels <300 ng/dl, i.e., levels currently considered diagnostic of hypogonadism [1]. Moreover, it is well established that serum total and free testosterone levels decline with aging in healthy men [2, 3]. Accordingly, the prevalence of low total testosterone levels increases in elderly men, affecting approximately 19, 28, and 49 % of men older than 60, 70, and 80 years, respectively [2]. However, only 27.7 % of men with low total testosterone levels have erectile dysfunction [1]. Moreover, serum total and free testosterone levels do not differ between patients with erectile dysfunction and age- and body mass index-matched healthy subjects [4, 5]. Indeed, even though androgens increase sexual desire, they do not play an important role in visually induced erections [6]. Accordingly, low testosterone levels do not appear to relate strongly to erectile dysfunction.

Despite the lack of a close relationship between hypogonadism and erectile dysfunction, measurement of serum total testosterone levels is currently recommended by most scientific societies in all patients with erectile dysfunction [7–9]. Among patients with erectile dysfunction, 4 % of those younger than 50 years and 9 % of those older than 50 years have low serum total testosterone levels [10]. In contrast, apparent signs of hypogonadism (i.e., small testes, gynecomastia, and reduced growth of body hair and beard) are rarely observed in patients with erectile dysfunction [11]. Indeed, if testosterone levels are measured only in patients with symptoms (e.g., low sexual desire) or signs suggestive of hypogonadism, 40 % of cases of low testosterone levels would be missed [10]. Total testosterone should be measured in the morning because serum testosterone levels exhibit a circadian variation and are higher in the morning [7]. Moreover, in patients with low testosterone levels, the diagnosis of hypogonadism should be confirmed by a second measurement, since as many as 30–40 % of patients will have normal levels on repeat testing [7, 8, 10, 12]. Testosterone measurements should not be performed in the presence of acute or subacute illnesses [7]. When total testosterone levels are near the lower limit of normal range and conditions that are associated with increased low sex-hormone-binding globulin (SHBG) levels are present, measurement of serum free testosterone levels might also be considered [7]. Common conditions that are characterized by higher SHBG levels and hence lower free testosterone levels include ageing, hyperthyroidism, and the use of estrogens, tamoxifen, and antiepileptic agents [7, 8]. It should emphasize that free testosterone levels should be measured only in reference laboratories because of the limited accuracy of many commercially available methods [7, 8]. In patients with documented low testosterone levels, serum luteinizing hormone, and follicle-stimulating hormone levels should also be measured to discriminate between primary and secondary hypogonadism [6–8]. In patients with secondary hypogonadism, measurement of serum prolactin levels and magnetic resonance imaging of the pituitary gland might be considered [7]. However, it should be mentioned that the prevalence of pituitary adenomas or hypothalamic lesions in these patients is low (3.1–6.7 %) [10, 13]. Accordingly, imaging might has to be reserved for patients with very low total testosterone levels (<100 ng/dl), panhypopituitarism, persistently elevated prolactin levels or symptoms suggestive of pituitary tumors (headache or visual field defects) [7].

Testosterone therapy can be considered in patients with erectile dysfunction who have low testosterone levels [7, 8]. However, it should be noted that administration of testosterone in placebo-controlled studies in patients with low testosterone levels exerted inconsistent effects on erectile function [14]. It is also unclear whether the combination of testosterone and phosphodiesterase-5 inhibitors improves erectile function in patients who do not respond to monotherapy with phosphodiesterase-5 inhibitors [7, 8, 15]. Several testosterone regimens are available (weekly or every 2 weeks im injections, patches, gel, pellets for sc implantation, and tablets for per os treatment), and the choice between them is individualized based on patient’s preference and cost [7]. After initiation of testosterone therapy, efficacy of treatment, serum testosterone levels (aiming at levels in the mid-normal range), and hematocrit and serum prostate-specific antigen (PSA) levels should be evaluated at 3–6 months and then annually [7]. Testosterone treatment is contraindicated in patients with prostate cancer or uncontrolled heart failure, whereas patients with PSA levels >3 ng/ml should be evaluated by a urologist before initiation of testosterone therapy [7]. Moreover, older patients are at increased risk for prostate-related adverse events (cancer, PSA >4 ng/ml, biopsy, urinary retention, or worsening of symptoms associated with prostatic hyperplasia) [16] and potentially for cardiovascular events when treated with testosterone [17, 18]. Accordingly, these patients may opt to avoid testosterone therapy [7].

Sexual dysfunction due to decreased libido or erectile dysfunction is the commonest presenting symptom in men with prolactinomas [19]. More than half of men with prolactinomas report symptoms of erectile dysfunction [20]. However, given the rarity of prolactinomas, only 0.2–0.6 % of patients with erectile dysfunction has prolactinoma [10, 21]. The pathogenesis of erectile dysfunction in prolactinomas is unclear. Most patients with prolactinomas who report erectile dysfunction also have testosterone deficiency, which might contribute to erectile dysfunction [21]. Elevated prolactin levels suppress the release of gonadotropin-releasing hormone and consequently the secretion of luteinizing hormone [6]. Moreover, primary hypothyroidism might also be present in patients with prolactinomas and might also play a role in the pathogenesis of erectile dysfunction [19]. Administration of cabergoline, a D₂ selective dopamine receptor agonist, improved nocturnal penile tumescence in patients with prolactinoma [20]. Interestingly, this favorable effect of cabergoline was observed only in patients who achieved normalization of prolactin levels, even when testosterone levels were not normalized [20].