Testosterone/estradiol ratio, is it useful in the diagnosis of erectile dysfunction and low sexual desire?

Abstract

Erectile dysfunction and low sexual desire are multifactorial diseases. The decrease in testosterone levels is one of the causes, but the effect of estradiol is not well known. Moreover, study has shown that the testosterone/estradiol ratio has more influence over sexuality than does estradiol alone. The aim of the study was to determine whether the balance between testosterone and estradiol has any relation to some aspects of sexual function. It was an ambispective study of 230 patients with urological problems unrelated to sexuality. They underwent a detailed history and hormone study including total, free, bioavailable testosterone and estradiol. They completed the Sexual Health Inventory for Men and questions 11 and 12 of the IIEF15 were used to assess impairment in sexual desire. The T/E ratio was calculated, and the relationship between the different parameters and erectile function and sexual desire were studied by univariate and multivariate analysis. The mean age was 66.32 ± 8.17 years. The percentage of patients with erectile dysfunction was 60.9% (7% severe, 14.3% moderate, 12.6% mild to moderate and 27% mild) and decreased sexual desire was 46.5%. Age, free and biodisponible testosteron were the only variables with a positive linear association with erectile dysfunction and decreased sexual desire. Age was the only independent variable for both, erectile dysfunction and sexual desire, in the multiple linear regression. There was no association between a testosterone/estradiol imbalance and an alteration in erectile function and sexual desire. Consequently, in the clinical study of these patients, it is not necessary to request estradiol in the laboratory analyses.

• Erectile dysfunction
• estradiol
• male sexual disorders
• sexual hormones
• testosterone

Introduction

Normal sexual function is the result of the confluence of psychological, neurological, hormonal and vascular factors. The last few decades have revealed several organic causes of erectile dysfunction (ED) that could account for up to 75–80% of these problems [1]. Many of them are directly related to the changes produced by the aging [2]. Additionally, the ingestion of various drugs and the presence of certain cardiovascular risk factors can substantially affect sexual functioning. There has been increasing interest in understanding these factors and the proper evaluation of them in clinical practice in order to offer patients better solutions to their problems. Among the factors to consider are hormonal changes. These can affect both erectile function and sexual desire [3]. Currently, recommended as good clinical practice is the study of certain hormonal parameters in patients with problems related to sexuality [4]. The descent in testosterone levels is one of the best known hormonal factors affecting sexual function [5,6]. Nearly, every component in EF is modulated by testosterone from pelvic ganglions to smooth muscle and the endothelial cells of the corpora cavernosa. It also regulates the timing of the erectile process, coordinating penile erection with sex [7]. However, the threshold of testosterone to maintain an erection is low and ED is usually a symptom of more severe cases of hypogonadism. For levels  >8 nmol/L the relationship between circulating testosterone and sexual functioning is very low.

Patients with low levels of testosterone may have erectile dysfunction and decreased libido. In the case of estrogens, it is not clear what role these hormones play as a modulating factor of sexual dysfunction in men. In recent years, it has been postulated that more than just the effect of estrogens themselves, erectile dysfunction and decreased sexual desire would be affected by the imbalance between testosterone and estradiol [8]. The aim of the present study was to assess whether there is an association between the testosterone/estradiol ratio (T/E ratio) and the alteration of sexuality (erectile function and sexual desire) in order to assess the usefulness of requesting a estradiol study in patients who present with this pathology.

Methods

Subjects and sample analysis

An ambispective analysis of 230 patients. These patients were included prospectively in a study conducted to evaluate male sex hormone changes during the ageing. Afterwards we decided to analyze retrospectively the same database and evaluate the association of the sex hormones (testosterone, estradiol and the T/E ratio) to common male sexual conditions of ED and diminished sexual desire.

The inclusion criteria was 50-year-old males who came to the clinic for a general urological disease or prostate review. The exclusion criteria were (a) any acute or chronic illness such as congestive heart failure, heart attack episode in the past year, neoplasic disease, brain stroke or severe osteoarthritis; (b) the use of concomitant medications or medical conditions that could affect hormone levels; (c) that the patient went to the clinic for erectile dysfunction.

All patients underwent a detailed history, including medications, lifestyle habits, a thorough physical examination and a complete blood analysis (blood biochemistry, blood count and hormonal determinations). The blood sample was taken between 8:00 and 10:00 am, and hormonal determinations included total testosterone (TT; Chemiluminescent Microparticle Immunoassay, Abbot Laboratories®), sex hormone-binding globulin (SHBG, DELFIA® SHBG kit, Wallac Oy, Turku, Finland), free testosterone (FT) calculated by the Vermeulen formula (TLC) and 17-estradiol (Auto DELFIA® Estradiol Kit, Wallac Oy, Turku, Finland).

Patients completed the Sexual Health Inventory for Men (SHIM) and questions 11 and 12 of the International Index of Erectile Function (IIEF-15) [9]. According to the SHIM results, patients were classified according to the severity of erectile dysfunction: severe erectile dysfunction patients with scores between 1 and 7, moderate between 8 and 11, mild to moderate between 12 and 16, mild between 17 and 25 and without erectile dysfunction if they had more than 25 points. Questions 11 (“During the last four weeks, how often have you felt sexual desire?”) and 12 (“During the past four weeks, how would you rate your level of sexual desire?”) of the IIEF-15 questionnaire were used to evaluate the sexual desire. Patients with scores between 1 and 5 were considered to have decreased sexual desire. Those with 6 to 10 points were considered to be without alteration.

To calculate the balance between testosterone and estradiol, the variable ratio T/E was created with the data of total testosterone and 17-estradiol obtained by laboratory tests.

Statistical analysis

A descriptive analysis of the sample and a univariate analysis (correlation for quantitative variables and comparing means for categorical variables) and multivariate (multiple linear regression) between the degree of erectile dysfunction and different clinical variables and laboratory parameters was conducted, and the same analysis was performed, with the dependent variable alteration of sexual desire. The different variables studied were age, diabetes, smoking, high blood pressure, waist circumference, BMI, total cholesterol, HDL and LDL fractions, triglycerides, total testosterone, free testosterone, bioavailable testosterone, 17-estradiol and the T/E ratio.

The statistical analysis was performed with SPSS 20 software and statistical significance was established as p ≤ 0.05.

Results

The mean age of the patients was 66.32 ± 8.17 years. Of these, 14.8% were diabetic, 28.7% had high blood pressure, 27.3% were current smokers, 13% were obese by BMI and 32% had abdominal obesity (abdominal perimeter ≥102 cm) [10]. After analyzing the results of the SHIM, 39.1% had no erectile dysfunction, 27% mild ED, 12.6% mild-to-moderate ED, 14.3% moderate ED and 7% severe ED. According to the results of questions 11 and 12 of the IIEF-15 questionnaire, 53.5% of patients had normal sexual desire and 46.5% of them had decreased sexual desire.

Table 1 shows the correlation studies of the SHIM score and sexual desire with quantitative independent variables. Age, free testosterone and bioavailable testosterones were the only variables related to the SHIM score (p < 0.05). For sexual desire, the related variables were age, bioavailable testosterone and free testosterone (p < 0.05). In addition, LDL cholesterol was related to sexual desire (p < 0.05).

Table 1. Descriptive and univariate analysis with pearson correlation for erectile dysfunction (SHIM score) and quantitative variables and sexual desire (items 11 and 12 of IIEF-15) and the same quantitative variables.

		Erectile dysfunction		Sexual desire
	Mean ± SD	r*	p	r*	p
Age (years)	66.32 ± 8.17	−0.548	0.000	−0.392	0.000
Total Cholesterol (mg/dL)	296.39 ± 32.69	0.082	0.237	0.102	0.141
HDL cholesterol (mg/dL)	52.07 ± 15.21	−0.092	0.188	−0.107	0.117
LDL cholesterol (mg/dL)	123.0 ± 31.21	0.134	0.053	0.027
Triglycerides (mg/dL)	110.20 ± 41.68	0.104	0.136	0.068	0.327
Abdominal perimeter (cm)	98.61 ± 7.19	−0.120	0.069	−0.62	0.355
BMI (Kg/m^2)	26.74 ± 2.70	−0.056	0.401	−0.20	0.760
Total testosterone (ng/mL)	5.12 ± 1.72	0.014	0.838	0.005	0.935
Free testosterone(ng/mL)	0.30 ± 0.10	0.283	0.000	0.179	0.007
Biodisponible testosterone (ng/mL)	6.79 ± 2.24	0.288	0.000	0.193	0.004
Estradiol (ng/mL)	0.026 ± 0.011	0.079	0.239	0.49	0.466
T/E ratio	0.014 ± 0.008	0.074	0.273	0.059	0.385

r*, Pearson’s correlation coefficient.

Regarding Table 2, no association between erectile dysfunction and qualitative clinical variables such as diabetes, hypertension or smoking, was observed (Student’s t-test). In the same way, there was no association found between sexual desire and the same variables.

Table 2. Descriptive and univariate analysis by Student’s t-test of binary variables and erectile dysfunction (SHIM score) and sexual desire (items 11 and 12 of IIEF-15).

		Erectile dysfunction (SHIM)		Sexual desire
	N (%)	Mean ± SD	p	Mean ± SD	p
HTA
No	164	17.74 ± 5.54	0.348	4.70 ± 1.42	0.707
Yes	66	16.94 ± 6.45		4.62 ± 1.55
Diabetes
No	196	17.68 ± 5.87	0.274	4.69 ± 1.68	0.793
Yes	34	16.50 ± 5.45		4.62 ± 1.63
Smoker
No	168	17.45 ± 5.73	0.790	4.60 ± 1.47	0.186
Yes	62	17.68 ± 6.08		4.89 ± 1.39

The results of the multiple linear regression model with erectile dysfunction as a dependent variable concluded that the only variable with independent power over dysfunction was the age of the patients (beta coefficient of −0.513 and p = 0.00). That is, the older the patients got, the lower the scores on the SHIM questionnaire and, in consequence, the greater the severity of ED. Similarly, when performing a multiple linear regression with sexual desire as the dependent variable, the only variable with independent character was age, (beta = −0.363 and p = 0.00). Consequently, the older patients had a lower sexual desire.

Discussion

Sexual dysfunction problems are commonly seen in the urology clinic, but for many people remain a taboo. In some cases, the patients themselves are unaware that the symptomatology of these disorders improves with proper treatment. In the present study, comprised of patients attending the clinic for problems unrelated to the sexuality, 60.9% had erectile dysfunction given the results of the SHIM questionnaire. These results are comparable to others in the literature, as in the epidemiological study of the Massachusetts Male Aging Study, with a global prevalence of ED of 52% in patients between 40 and 70 years old, but reaching 67% in patients 70 years of age [11].

One of the indispensable complementary tests in the study of erectile dysfunction is the determination of testosterone blood levels [12]. Testosterone plays a key role not only in modulating erectile function at the central level and in the corpora cavernosa endothelium, but also in the coordination of sexual desire and erection [7]. In some representative samples of hypogonadism, prevalence values of erectile dysfunction of upto 23–36% were obtained [13], depending on the variance of the testosterone levels as a proposed cutoff. Patients with erectile dysfunction and testosterone levels below the lower limit benefit from this hormone replacement therapy, thus demonstrating that testosterone is a well-recommended treatment for the problem and allows for the prescription of a best treatment for each patient [14]. Decreased sexual desire also presents a clear association with decreased testosterone levels. In the study published by Crown in 2013, hypogonadism was found to be a risk factor for secondary decreased libido [15]. Similarly, in the study by Cunningham et al., the levels of total testosterone and free testosterone had a positive and independent association with erectile dysfunction and decreased libido in 788 patients over 65 years of age with low levels of testosterone [16]. In the present study, total testosterone did not appear as a variable associated with erectile dysfunction or decreased sexual desire; however, only if the variables have calculated free testosterone and bioavailable testosterone. Other authors, such as Ahn et al., have obtained similar results of association between free testosterone, but not with total testosterone levels [17]. As noted in previous studies, this could be justified by the fact that the frequency of biochemical hypogonadism is higher when quantifying free testosterone levels rather than those for total testosterone [18].

The role of estrogens in sexually reproductive males is not well defined. In the study by O'Connor et al., an association between both testosterone levels and ED, as well as with the frequency of masturbation was observed, but this association was not demonstrated with estradiol. Estradiol levels were, however, associated with discomfort in sexual function, as measured by the EMAS-Sexual Function Questionnaire [19]. Even after adjusting for testosterone levels, it was observed that those men with higher levels of estradiol had more discomfort in their sexual function according to the survey results cited above. According to their hypothesis, two different mechanisms of action in sexual function could have an affect: testosterone levels that influence behavior, and estradiol levels that decreased the sexual function associated with a depressive state. Recent hypotheses have given more importance to the imbalance that occurs in aging [20] between testosterone and estradiol levels more than estradiol levels alone. Thus, with increasing age, testosterone aromatization to estradiol, along with low testosterone levels, could be one of the causes of abnormal sexual function in men. Several studies in animal models by Srilatha and Adaikan, an increased intake of estrogens and phytoestrogens in experimental animals caused an increase in estrogen levels thus leading to an imbalance between the levels of testosterone and estradiol and triggering erectile function and other types of impaired sexuality [21,22]. Vignozzi et al., in their study conducted in rabbits, which were induced a metabolic syndrome due to a diet rich in fats, noticed that there was an increase in the estradiol levels and a decrease in the testosterone levels. Only the increase in the estrogen levels was associated independently with the rabbits erectile dysfunction problems. These inducted erectile problems improved partially with the supply of exogenous testosterone, as well as with the supply of antiestrogens like tamoxifen. In this study and in the animal models mentioned before, levels of hyperestrogenism are induced [23]. The results are not comparable with the ones obtained in our study, in which the patients do not have any exogenous supply of estrogens and do not have either any situation that induce them.

In the study by El-Sakka in 2013, the hormone levels of patients with erectile dysfunction were studied. The results obtained were that the severity of erectile dysfunction was independently associated with low testosterone levels, high levels of estrogens and a combination of both [24]. Emanuela Greco et al., [25] in their pilot study to assess changes in serum levels of sex hormones after 12 months of treatment with tadalafil in patients with erectile dysfunction, found an improvement in the responses to the IIEF-15. Patients had a decrease in serum estradiol levels, but an increase in testosterone levels, resulting in an increase in the T/E ratio. They postulated that this could have an inhibitory effect on aromatase that would produce an improvement in sexual function. In the present study, however, no association between estradiol and the parameters of sexual function studied were observed, and no association between the T/E ratio and the severity of erectile dysfunction and decreased libido was obtained, even after adjusting by age. In a study by Robert S. Tan et al., in which 34.016 patients treated with exogenous testosterone and with levels of hyperestrogenism (due to aromatization of testosterone to estradiol in adipose tissue), showed no association between high estradiol levels and decreased sexual desire. To the contrary, decreased libido was associated with normal and low estradiol levels [26]. Even in the longitudinal study of Benjamin et al., in which the hormonal profile of patients over 70 years of age for two years was studied, an association between estrogen levels and impaired sexual function in elderly men was obtained. There was an association between low testosterone levels and decreased sexual activity and desire, but not ED [27]. In the same study, the authors argued that their results could be that low levels of testosterone were not the cause of decreased sexual activity and desire, but it was a decreased sexual activity that was the cause and not the effect of presenting somewhat lower testosterone levels. Wu et al, in their study with four patient groups (control group, DE group, premature ejaculation group and premature ejaculation and erectile dysfunction group) noticed that the DE patient group had the highest levels of estrogens and normal levels of testosterone, compared with the control group. However they submit that the normal/average levels of testosterone may be due to the age of the patients, whose mean age were 37, 14 and 35, 54 years, respectively. They conclude that the results may show a clinical evidence to the etiology of erectile dysfunction in earlier ages [28]. However, these conclusions cannot be extrapolated to patients of older ages, as the ones in the sample of our study.

Age is the only variable in the present study that gives an independent association for both impaired erectile function and sexual desire. This result makes sense, since erectile dysfunction and decreased sexual desire are more prevalent with age. Many epidemiological and population studies have well shown that age is the most important independent factor [29,30]. Age is associated with a number of comorbidities and risk factors that affect sexuality, including neurological disorders, vascular dysfunction, decreased hormones and the side effects of certain medications. However, when we adjust for comorbidities, age continues to act as an independent predictor of erectile dysfunction and low sexual desire, probably due to tissue aging occurring with age and other factors that have not been analyzed in this study.

The present work has some limitations. It is an ambispective study with the design errors that this may entail. Patients who underwent the study were patients attending the clinic for urological problems unrelated to sexuality, so this may not be a fully representative sample of the general population. Estrogen levels in the patients in our sample were within the normal range, so the results cannot be extrapolated to a population that present levels of hyperestrogenism whose results may vary.

Conclusions

Despite the potential limitations of the study, and in accordance with the results, we can conclude that it is not necessary to include estradiol in an hormonal study in daily clinical practice in patients who consult their urologist on issues of sexual function, both erectile dysfunction and decreased libido. The role of the imbalance between estrogen, testosterone and estradiol in relation to the alteration of the sexuality has yet to be determined. Further studies are needed in order to help understand their role in this disease.

Declaration of interest

All the authors report no conflicts of interests. The authors alone are responsible for the content and writing of this article.