A Review on Testosterone: Estradiol Ratio—Does It Matter, How Do You Measure It, and Can You Optimize It?

Abstract

There is a natural balance between the major sex steroids, testosterone and estradiol, controlled by gonadal secretion and peripheral conversion by aromatase. This balance is impacted by a variety of inborn and acquired conditions, and, more recently, by a growing use of exogenous testosterone therapy and off-label aromatase use under the guise of “men’s health.” We summarize reported testosterone:estradiol ratios, both naturally occurring and with pharmacologic manipulation and consider the ramifications of significant changes in these ratios. However, significant limitations exist in terms of steroid separation and measurement techniques, timing of samples, and lack of consistency from one assay to another, as well as definition of normative data. Limited data on the testosterone:estradiol ratio in men exists, particularly due to the scan data on concurrent estradiol values in men receiving testosterone therapy or aromatase inhibitors. Nonetheless, there seems to be a range of apparently beneficial values of the testosterone: estradiol radio at between 10 and 30, calculated as: testosterone in ng/dL/estradiol in pg/mL. Higher values appear to be associated with improved spermatogenesis and reduced bone density while lower values are associated with thyroid dysfunction. While there is growing awareness of the significance of the testosterone:estradiol ratio, and a sense of a desired range, the optimal value has not yet been determined. Further work is needed to clarify the measurement strategies and clearly-defined outcome measures related to the testosterone:estradiol ratio.

INTRODUCTION

Testosterone and estradiol, the primary sex steroids, are present in both men and women, at different amounts over the course of their lives. Both are necessary for optimal health and function. The balance of these hormones shifts over time, including the overall lifespan, the years of active reproduction, as well as over monthly (in women) and daily cycles. So, while the ratio of testosterone to estradiol varies over these different cycles, there is a natural ebb and flow. Recent pharmacologic developments have the potential to impact this natural state, and we were interested in exploring the potential effects of these alterations primarily, but not exclusively, in men. A recent review by Russell and Grossmann elegantly addresses the role of estradiol as a male hormone [1]. We will examine this role of estradiol by considering its relationship to its precursor, testosterone, as expressed in the T:E ratio. We believe that many current studies exploring modulation of testosterone in men’s health ignore the key role of estradiol and overlook the relationship between the major sex steroids. In this review, we hope to correct this perception. Unfortunately, in our review of the literature, we could find no study exploring the overall relationship of the T:E ratio and health outcomes.

MATERIALS AND METHODS

The authors searched PubMed between September 2022 and December 2023 for English-language articles using the search terms, “testosterone:estradiol ratio, testosterone measurement, estradiol measurement, aromatase, estradiol:testosterone ratio, and men’s health clinics.” In addition, a Google search focusing on “men’s health” was used to identify both information available to the public as well as information on clinical practices of the providers in this area. Finally, literature from the authors’ personal libraries and resources were reviewed. This article used published data only and no new studies in humans or animals were performed.

For consistency, unless otherwise stated, the testosterone:estradiol ratio is derived from total testosterone in ng/dL and total estradiol in pg/mL, and expressed as T:E.

TIMELINESS OF THE TOPIC

This interest is timely for several reasons:

1. Growing awareness of conditions affecting the normal T:E ratio

These conditions primarily reflect endogenous aromatase activity. Although later in this discussion we will be reviewing pharmacologic manipulation of aromatase (primarily inhibition), we are here addressing naturally occurring variation in aromatase activity. Aromatase carries out the conversion of androgen to estrogen. This enzyme is present in multiple organs including adipose tissue, brain, blood vessels, skin, bone, endometrium, and breast tissue in both sexes [2,3,4,5,6]. The human aromatase enzyme is a member of the cytochrome P450 family and is the product of the CYP19A1 gene, located on chromosome 15 [7,8]. It functions to catalyze the rate-limiting and final step of estrogen biosynthesis; the aromatization of androgens to estrogens. It does this via three oxidation reactions of the androstenedione A ring, with each reaction consuming a molecule of both oxygen and nicotinamide adenine dinucleotide phosphate (NADPH) per reaction. Of these three steps, the third is unique to aromatase, while the first two are common to P450 cytochrome proteins [9]. Estrogens are known to be important in the growth of breast cancers in both pre- and post-menopausal women. When ovaries are no longer functional, the source of estrogens in postmenopausal women comes from the peripheral conversion of androgens (produced in the adrenals and peripheral tissue) by the aromatase enzyme. Breast cancer tissues have been shown to express aromatase and produce higher levels of estrogens than non-cancerous cells. This is one of the main reasons that aromatase inhibition has generated a high level of interest for treatment of breast cancer [10]; we will examine later the role of aromatase inhibition in the treatment of hypogonadal men.

2. Pharmacologic manipulation of circulating testosterone values

There is a proliferation of “men’s health” clinics (usually, but not always, for-profit) using various combinations of testosterone supplementation, aromatase inhibitors (AIs) (which inhibit the conversion of testosterone to estradiol), 5-α-reductase inhibitors (which inhibit the conversion of testosterone to dihydrotestosterone), selective estrogen receptor modulators (like clomiphene and other agents), and gonadotropins like human chorionic gonadotropin (like HCG) with the goal of boosting total and free testosterone, purportedly to improve libido, sexual performance, muscle development, and, potentially, to reverse age-related changes (e.g., hair loss, prostate growth). Indeed, some practitioners promote the use of testosterone with a focus on minimizing the possible effect of age on muscle mass and strength, sexual performance, etc. This marketing is geared to individuals concerned with the age-related decline in testosterone in men (with rare attention paid to the concomitant reduction in estradiol production). While these clinics have, to some extent, expanded to address male endocrine, urologic, physical performance, and psychological issues, many began, and remain, as testosterone therapy centers [11]. It should be pointed out that most professional endocrine societies strongly recommend that testosterone therapy be offered to men with symptomatic testosterone deficiency (in whom the diagnosis has been confirmed by repeated measurements of total, and in some cases, free, testosterone as well as clinical symptoms) to induce and maintain secondary sex characteristics after a discussion of potential benefits and risks of therapy [12,13,14].

There is a fine line between use, misuse, and abuse; those distinctions are beyond the scope of this article except to highlight the focus on boosting testosterone [15,16,17,18,19]. Indeed, some authors have commented on questionable marketing strategies [20], with potentially minimal validation of a patient’s need for testosterone and adherence to recognized standards of care [21,22].

An obvious example of the manipulation of circulating testosterone is in the treatment of patients with gender dysphoria. While not all patients who perceive gender incongruence (a term used when gender identity and/or gender expression differs from what is typically associated with the assigned gender [23]). In these patients, gender-affirming therapy varies in terms of the desired degree of gender expression, and as such is highly individualized. These therapeutic strategies involved both suppressing endogenous sex steroid production/signaling as well as providing exogenous sex steroid for the desire expression. Treatment goals depend on both clinical and laboratory markers of gender affirmation. Because of the wide variation in treatment strategies, rigid goals for circulating sex steroid concentrations do not apply, but in general, for trans-women, testosterone should be <50 ng/dL, and estradiol should be between 100 and 200 pg/mL. For trans-men, testosterone should be in the normal adult male range (400–700 ng/dL). Estradiol concentrations in trans-men are not specified [23].

3. Increasing off-label use of aromatase inhibitors

Currently, AIs that are now in clinical use and are approved by the US Food and Drug Administration (FDA) for the treatment of breast cancer in postmenopausal women include anastrozole, letrozole (both “nonsteroidal AI”), and exemestane (a “steroidal AI”—this nomenclature is based on the structural similarity of the inhibitor molecule to the androgen substrate, which in postmenopausal women, is typically androstenedione). AIs are approved for postmenopausal women with hormone receptor-positive breast cancer in both the adjuvant and metastatic setting [24], and, while side effects are not uncommon, these rarely lead to discontinuation of treatment [25]. Thus, despite their use in gynecologic malignancy, uses in other clinical contexts are non-FDA-approved, or “off-label.” Aromatase is a target for inhibitor therapy in estrogen-sensitive diseases such as endometriosis and leiomyoma, in addition to cancer. However, all tissues contain estrogen receptor-expressing cells; there are a multitude of non-receptor effects of estrogens, and so it could be anticipated that the effect of aromatase is universal. Of note, it has been reported that males have two- to three-fold greater the aromatase activity of females [26].

Among the off-label uses of AI is improving male fertility, e.g., in patients with Klinefelter’s syndrome, as well as men with other causes of infertility [27,28], to improve sperm production, as well as improving yield from surgical sperm extraction [29]. A meta-analysis by Del Giudice et al [27], reported a dramatic increase in total testosterone (from 320 to 476 ng/dL) and the T:E ratio (from 7.4 to 24.1) and a drop in estradiol (from 37 to 27 pg/mL). However, there are potentially unintended consequences of any treatment. Thus, efforts to increase testosterone, at the expense of estradiol, using AIs have been associated with polycythemia [30], osteoporosis [31], transaminitis [32], and decreased libido (which may defeat the purpose of taking the AI in some cases [27]).

Independently of the impact of AIs, endogenous aromatase activity reflects other factors as well. In addition to obesity, which may affect total aromatase activity, dietary components such as red wine (reduce) and vitamin D (increase), may impact aromatase activity [31]. The effect of aromatase is reflected in the “threshold estradiol hypothesis for skeletal sufficiency in the male which suggests that a threshold level of bioavailable E is needed for skeletal E sufficiency in the male, and as men fall below this level, the risk of osteoporosis and bone loss increases [33].” This suggests a role for estradiol, and its measurement, beyond total and free testosterone. Overall, anastrozole is not recommended as a routine treatment for low testosterone [34].

In addition to interfering with the aromatization of testosterone, many clinicians attempt to boost testosterone by interfering with its reduction to another androgen, dihydrotestosterone. Although one would expect that interfering with this conversion by inhibiting 5-α-reductase would increase testosterone concentrations, the results are inconsistent, as pointed out in a meta-analysis by Traish et al [35]. Nonetheless, agents like finasteride and dutasteride are widely used. It may be that 5-α-reductase inhibitors are used not so much for increasing testosterone as reducing the side effects of dihydrotestosterone in terms of prostatic enlargement and androgenic alopecia. Nevertheless, there is a slight increase in testosterone in men with low baseline values; as stated, results in other groups are equivocal [35]. The effect on the T:E ratio is unclear, with some studies showing no effect on estradiol [36], some showing an increase in production of estradiol [37], some showing a reduction in estradiol in lean, but not obese subjects [38], and some suggesting an aromatase-independent inhibition of estrogen synthesis [39].

4. The increasing burden of obesity

Obesity contributes to low testosterone in several ways. As stated above, aromatase is present in a variety of tissues, including adipose tissue. Thus, in obesity, whole-body aromatase activity is increased (even if percell aromatase activity is reduced in mature adipocytes compared to pre-adipocytes [31]), likely explaining the increase in testosterone conversion to estradiol seen in obese men [40]. Other explanations for the low testosterone in obesity include decreases in total testosterone due to insulin resistance-associated reductions in sex hormone binding globulin. More severe obesity is additionally associated with reductions in free testosterone levels due to suppression of the HPT axis. Low testosterone by itself leads to increasing adiposity, creating a self-perpetuating cycle of metabolic complications [41]. A recent study by Ciardullo et al [42], using data from the National Health and Nutrition Examination Survey (3,309 men were studied), noted that when the participants were stratified by total testosterone quartiles, there was a stepwise increased in the T:E ratio, from 11.2, to 15.8, to 18.6, to 22.8.

5. A growing older population “aging boomers”

Testosterone has become one of the most widely prescribed medications in the USA, increasing consistently according to a 2023 analysis [43]. This increase has resulted in the dramatic growth of the testosterone therapy sector of the pharmaceutical industry from US $18 million in the 1980s to US $1.6 billion in 2011 [17]. There was a slight but apparently transient drop in testosterone use in 2016 but usage has increased since [44]. The reason is multifactorial but can partly be attributed to the continued growth of the population over 65 years of age and a greater awareness of medical comorbidities more prevalent with age and associated with low testosterone, such as metabolic syndrome (MetS) and cardiovascular disease (CVD) [45,46,47].

While the use of hormone replacement therapy with estrogen in menopausal women is beyond the scope of this article, recent evidence supporting concomitant testosterone supplementation in these women has been addressed; the benefits appear to be limited to enhancement of sexual function and are not universal [48]. Indeed, the Endocrine Society recommends against routine use of androgen in menopausal women except in cases of hypoactive sexual desire disorder where short courses can be used [49]. Nonetheless, the use of exogenous estrogen and testosterone potentially could impact the T:E ratio in these women. In addition, in premenopausal women, the phase of the menstrual cycle impacts bioavailable testosterone and estradiol and muscle strength [50].

6. Increasing burden of osteoporosis

Not surprisingly given an aging population, the prevalence of osteoporosis is increasing. A recent meta-analysis, using population-based studies based on World Health Organization criteria, estimates that the global prevalence of osteoporosis and osteopenia was 19.7% (95% confidence interval [CI], 18.0%–21.4%) and 40.4% (95% CI, 36.9%–43.8%) [51]. While estradiol is recognized as having significant benefit for bone health (elaborated as the “threshold estradiol hypothesis” described previously [33]), efforts to link this tissue-level benefit to circulating testosterone and estradiol values have been unsuccessful. Yeap et al [52] report based on a study of incident fracture in over 3,000 elderly men, that mid-range plasma testosterone was associated with lower incidence of any fracture and hip fracture, and higher sex hormone-binding globulin (SHBG) with increased risk of hip fracture and suggest that circulating androgen rather than estrogen represents a biomarker for hormone effects on bone driving fracture risk. Similarly, Orwoll et al [53], reporting from the multicenter Osteoporotic Fractures in Men Study Consortium, describe a limited clinical utility of serum estradiol, testosterone, and SHBG measures for the evaluation of osteoporosis risk in elderly men. These authors reported total and bioavailable testosterone and estradiol; our calculation of T:E ratios ranged from 19.8 to 21.9 for total sex steroids, and 16.5 to 17.4 for bioavailable hormone. A provocative recent case report by Lanfranco et al [54] describes a 26-year-old man with a mutation in the aromatase gene who presented with tall stature, eunuchoid proportions, unfused epiphyzes and osteopenia; estradiol was undetectable. Genetic analysis revealed a point mutation in the CYP19A1 gene. Bone markers improved after estradiol was raised above 73 pmol/L (19.9 pg/mL [authors’ calculation]), with concurrent total testosterone at 599 ng/dL.; resulting in a T:E of 30.1 (authors’ calculation). Further evidence for the role of estradiol in the regulation of bone mass in men comes from a genome-wide association study using samples from 11,097 men in nine European cohorts, which demonstrated that variants in the aromatase gene affecting estradiol and estrone levels, impacted bone health in men, highlighting the importance of estradiol; however, testosterone and estradiol values were not reported [55]. Jamall et al [56] describe a man with osteoporosis and an aromatase mutation, whose bone density only improved when he was started on testosterone, with an increase in circulating estradiol as well as testosterone, and his estradiol concentration exceeded 20 mg/mL (baseline of 8). Of note, while testosterone also increased, the T:E ratio actually declined from about 30:1 to 20:1, suggesting a threshold effect of estradiol on bone health. Similar observations reinforcing the role of an estradiol threshold have been reported by Rochira [57] and Mellström et al [58].

7. Gynecomastia

Breast enlargement in men can be socially awkward as well as uncomfortable. It typically occurs in puberty, with aging, medications, and health conditions that affect hormones. Almost 40 years ago, Pont et al [59] reported an alteration in the T:E ratio in patients taking ketoconazole, with higher estrogen associated with gynecomastia. More recent studies have supported this conclusion, whether the gynecomastia is associated with medication, or underlying conditions [60,61]. Unfortunately, aromatase inhibition seems to be ineffective at reversing gynecomastia [62].

8. Hypothalamic-pituitary-testicular axis regulation

Studies in animals and humans demonstrate that estradiol acts indirectly in the hypothalamus to regulate gonadotropin-releasing hormone (GnRH) neurons [63], and directly in the pituitary to affect gonadotropin production [64,65]. Estradiol may also act directly on the pituitary to reduce sensitivity to GnRH [65]. Studies of the AI anastrozole in normal men compared to men with hypogonadotropic hypogonadism demonstrated that aromatization of testosterone to estradiol is required for negative feedback of LH secretion. The authors speculate that estrogen acts at the hypothalamus to decrease GnRH pulse frequency and at the pituitary to decrease responsiveness to GnRH [66]. In men with Klinefelter’s syndrome, while gonadotropins were higher than in controls; estradiol was slightly, but not significantly higher than in controls. Importantly, the T:E ratio (calculated by us) was 10.4 (lower than in controls [67]), suggesting a major role of estradiol in gonadotropin regulation.

9. Growth hormone-IGF-1 axis regulation

Growth hormone (GH) is traditionally considered to be the main regulator of growth. During puberty, elevated sex steroid concentrations (especially estrogens) stimulate GH production, leading to an activation of the whole GH/insulin-like growth factor-1 (IGF-1) axis [68]. This relationship is exemplified by the observation that men with congenital aromatase deficiency had impaired GH secretion and low IGF-1 [69]. These men had low baseline IGF-1, and an impaired GH response to combined GHRH-arginine stimulation. Unfortunately, exogenous estradiol in these men did not normalize GH secretion, suggesting a persisting effect of congenital estradiol deficiency or a role of local (intrahypothalamic) estradiol production. The authors did not report circulating testosterone or estradiol values, and so the T:E ratio cannot be calculated. Birzniece and Ho [70] have emphasized the role of sex steroids in supporting the GH axis, in particular the effect of local aromatase and a paracrine effect. In addition, exogenous estrogen stimulates GH secretion, but can inhibit GH action when given orally; suggesting an important role for non-oral estrogen administration; again no circulating values for testosterone or estradiol were reported. In a study of testosterone-replaced older hypogonadal men, estradiol addback during aromatase inhibition increased basal, pulsatile, and total GH secretion. At baseline, testosterone in the study subjects averaged roughly 430 ng/dL and estradiol about 22 pg/mL, for a roughly-calculated T:E of 19.5 or so [71]. In aggregate, the data suggest that estradiol rather than testosterone is the primary sex steroid regulator of the GH-IGF-1 axis in men.

EFFECTS OF CHANGES IN T:E

Clearly, the T:E ratio can be impacted by a variety of factors and conditions, some occurring naturally, and some in response either to administration of exogenous sex steroids or manipulation of their conversion or production. In addition to the pharmacologic perturbation of the T:E ratio, there are naturally occurring factors at play. As a result, there is observational data on alterations of the T:E ratio. There are few studies looking at the T:E ratio and its effect on health. In a retrospective study based on three NHANES (National Health and Nutrition Examination Survey) cycles, Belladelli et al [72] reported a significant association between low testosterone (<250 ng/dL) and overall mortality, and a low T:E ratio (calculated as testosterone [ng/dL]/estradiol [pg/mL] <10) and CVD-related mortality. In addition, an increase in the T:E ratio (calculated by the authors of this paper and based on testosterone in ng/dL/estradiol in pg/mL as rising from 15.9 to 16.7) has been associated with reduced bone mineral density [73]; a similar calculation by us showed the progression in T:E from 12.4 to 26.2 associated with a risk for osteoporosis [74]. While provocative, these findings must be interpreted with findings reviewed above showing modest benefit of circulating testosterone and estradiol on bone health [52,53].

In a prospective, office-based study, El-Sakka [75] reported that while a low testosterone (<280 ng/dL) had a significant effect on erectile function, a concomitant elevated estradiol value (>42.6 pg/mL) further impaired erectile function. Again, we have calculated this T:E ratio for erectile function at 6.57. In contrast, Castelló-Porcar and Martínez-Jabaloyas [76] reported no association between the testosterone/estradiol ratio and erectile function or desire in Spanish men over the age of 50 years.

As stated above, aromatase inhibition has been associated with improved fertility; Schlegel [77] reports that testolactone use results in the T:E ratio increasing from 5 to 12, and anastrozole use increasing the ratio from 7 to 18, with a corresponding increase in sperm concentration, motility, and morphology. Similar findings were reported by Shoshany et al [78], with an increase in the T:E ratio from 6.98 to 34.5 at 3 weeks of treatment; this improvement was sustained, along with improved sperm parameters. Moreover, Esteves [79] notes how the T:E ratio can affect the treatment approach of men with spermatogenic failure leading to nonobstructive azoospermia. This author points out that approximately half of these men will have low testosterone (<300 ng/dL), and a number will have elevated estradiol (>60 pg/mL), especially if they are obese. Measurement of sex steroids, as well as SHBG, may thus be helpful in decision making in terms of identifying candidates for medical therapy, such as HCG, clomiphene, or AIs. However, these measurements must be tempered with the wide range of values that can be obtained as well as global geographic variations in test availability [79].

On the other hand, a decrease in the T:E ratio has been linked to increases in markers of autoimmune thyroid disease in Chinese men, specifically serologic (thyroid peroxidase antibody and/or thyroglobulin antibody) and ultrasound (ultrasonographic diffuse parenchymal hypoechogenicity and/or a heterogeneous echogenic pattern). These authors reported their sex steroid data as log-transformed E (in pmol/L):T (in nmol/L); 1.86–1.94 compared to 1.77 for increased risk of autoimmune thyroid disease [80]. As elsewhere, for purposes of comparison in this review, we have recalculated these values in mass, rather than molar, units.

However, it is important to recognize the importance that both androgen and estrogen signaling have on normal physiology. A randomized trial conducted by Finkelstein et al [81] where reproductive aged men were chemically castrated and then randomized to increasing doses of transdermal testosterone therapy with or without concomitant aromatase therapy demonstrated that lean mass, muscle size, and strength are regulated by androgens whereas fat accumulation is primarily due to estrogen deficiency. Moreover, the authors concluded that sexual function is regulated by both androgens and estrogens. Thus, the balance of both hormones is important.

LIMITATIONS OF THE T:E RATIO: HOW TO MEASURE T:E?

Because of the many factors affecting both testosterone and estradiol concentrations, measuring these hormones is not a trivial task. Currently, only immunoassay and mass spectrometry methods are suitable for clinical applications. However, applying these methods to biological specimens like blood requires isolation of the steroids both from other components of the specimen and other steroids, with extraction and chromatographic methods, prior to immunoassays or mass spectroscopy measurement techniques [82]. In terms of testosterone measurement, The Endocrine Society recommends measuring total testosterone in men in the morning (times not specified [although VA recommends between 8 and 10 a.m.] in the fasting state), with an “accurate and reliable assay” [12]. A currently unresolved problem is that the reported reference ranges for total and free testosterone vary widely among laboratories and assays; this is particularly an issue with measuring free testosterone [12]. In addition, most of the current assays are less accurate with testosterone concentrations at the lower end of the normal range for men, and across the entire range for women. Accuracy can be improved, in men, with fasting morning samples, and, in premenstrual women, measurement of testosterone should be performed at the follicular phase [83].

Measurement of free testosterone is more challenging than measuring total; a recent article by Fiers et al [84] reports that free testosterone by equilibrium dialysis is reliable and may be best approximated by use of the Vermuelen equation. Some authors have proposed measures of bioavailable testosterone and estradiol, but these create other technical problems since these are measures of albumin-bound hormone, which is affected by aging and nutritional status [85,86,87].

Of note, AIs may also impact sex hormone binding globulin, affecting both total and free testosterone measurements. Exemestane significantly suppresses sex hormone binding globulin [88]. A study in women being treated for breast cancer showed that AI lowered SHBG and increased free testosterone [89]. On the other hand, a study in subfertile hypogonadal men showed no difference in SHBG with anastrozole treatment [89]

Measurement of salivary sex steroids, in analogy to cortisol measurement, is relatively new [90,91], and may be of value in adolescent boys and men but is not informative at low levels [91] and thus may be less useful in studying women.

The measurement of estradiol has also been challenging. The wide variation in estradiol measurements occurring naturally over a woman’s life, as well as those that obtain from pharmacologic manipulation, impose great demands on an assay’s performance; to date, routine clinical assays fall short in a number of areas [82].

The Table 1 shows some of the T:E ratios that have been reported. We have tried to identify the assay method and the temporal components, when possible. It is apparent that different outcomes have optimal T:E ratios.

Table 1. Clinical manifestations of the testosterone:estradiol (T:E) ratio^a.

Sex steroid ratio	Hormone assay method	Outcome	Reference
T:E <10	Electrochemiluminescence	Increased risk of CVD-related mortality	Belladelli et al [72]
Log-transformed E (in pmol/L):T (in nmol/L); 1.86–1.94 compared to 1.77 for increased risk of autoimmune thyroid disease	Chemiluminescence	Thyroid antibody positive, US features of autoimmune thyroid disease	Chen et al [80]
Variable; groups stratified into low, medium and high estradiol and testosterone; T:E 15.9 for no fracture, 16.7 for fracture	Radioimmunoassay	Increased risk of hip fracture with low estradiol, especially with low testosterone	Amin et al [74]
19.8–21.9 for total T:E	Sex steroids were measured using gas chromatography/mass spectrometry, and SHBG by radioimmunoassay	No effect on osteoporosis risk	Orwoll et al [53]
Variable; groups stratified into low, medium, and high estradiol and testosterone. Cutoff 6.57	Electrochemiluminescence immunoassays	Erectile function impacted by low testosterone; elevated estradiol had an additive effect	El-Sakka [75]

This table summarizes published literature addressing some aspects of the testosterone: estradiol ratio, including reported (or calculated ratios), how the hormones were measured, and reported clinical outcome, if any.

SHBG: sex hormone-binding globulin, CVD: cardiovascular disease.

^aUnless otherwise stated, calculated as total testosterone in ng/dL, total estradiol in pg/mL.

CONCLUSIONS

There are gaps between conclusions that can be drawn from the available data, the quality of that data, and where to go from here.

While we have focused on the T:E ratio and men’s health, a useful perspective can be found in a specialized area of women’s’ health: Polycystic ovary syndrome (PCOS). For example, Amato et al [92] have pointed out that a reduced E:T ratio (comparable to an elevated T:E ratio), measured in the follicular phase and reported as estradiol (pmol/L): testosterone (nmol/L) is associated with oligo-anovulatory cycles and atherogenic lipid profile: normal women had an E:T ratio of 141.9, PCOS patients with normal cycles had a ratio of 96.4, while those with oligo-anovulatory cycles had a ratio of 75.4. For consistency in this review, we have converted (using the online calculator at unitslab.com) these data to pg/mL for estradiol (using the conversion factor 1 pg/mL=0.272405 pmol/L) and ng/dL for testosterone (using the conversion factor 1 ng/dL=28.818444 nmol/L) and rearranged the data as T:E ratios. When these calculations are performed, normal women had a T:E ratio of 0.76, PCOS women with normal cycles had a ratio of 1.19, and PCOS women with oligo-anovulatory cycles had a ratio of 1.44 (i.e., nearly a two fold increase from normal). Franik et al [93] report in a single-center study that estradiol:testosterone and estradiol:androstenedione ratios were lower in PCOS than non-PCOS subjects (estradiol, testosterone, and androstenedione were measured with enzyme-linked immunosorbent assays [ELISA] in the follicular phase and reported as pg/mL, ng/mL, and ng/mL respectively), but did not differ significantly between obese and normal-weight groups. Indeed, the lowest values were in the normal-weight PCOS subjects, suggesting a role for nutritional status in aromatase activity: in non-obese subjects, the E2:T ratios were 58.5 compared to 96.5 in PCOS and non-PCOS women (testosterone reported as ng/mL and estradiol as pg/mL), respectively, while the estradiol:androstenedione ratios were 20.7 and 24.2, respectively. Rearranging the data to express testosterone as ng/dL and estradiol as pg/mL, we get T:E ratios of 1.61 and 1.35 in non-obese and obese PCOS women, respectively, and 1.04 and 1.03 in normal weight and obese non-PCOS women, respectively, similar to those reported by Amato et al [92].

So, we have seen, in very general terms, that the T:E ratio has significance in terms of overall health as well as specific areas of concern for both men and women, that ratio, as presently calculated from the data, is broad and useful only in the most general terms. The data is impacted by inconsistencies in methods of steroid isolation and measurement, as well as timing of sampling, and symptom reporting. Future studies should attempt to synchronize methodologies to improve interpretability. It might then be possible to apply the resulting understanding of T:E ratio to other situations, such as men receiving testosterone supplementation, and individuals receiving aromatase therapy. Ultimately, randomized controlled trials exploring the “ideal” T:E ratio would be desirable, but it may be that different “goal” ratios would be necessary for fertility, thyroid function, erectile function, bone health, etc. In addition, it appears that the T:E ratio is further affected by ambient estradiol concentrations, suggesting a “threshold” effect. Finally, it is important to consider the paracrine role of local aromatase activity in the hypothalamus and pituitary, resulting in local conversion of testosterone to estradiol not reflected in circulating values.