Significant association between serum dihydrotestosterone level and prostate volume among Taiwanese men aged 40–79 years

Abstract

Introduction. We evaluated the association between serum sex hormone levels and prostate volume in Taiwanese men.

Methods. A cross-sectional study was conducted in 505 men (aged 40–79 years, mean age 58 years). Serum total testosterone (TT), free testosterone (FT), dihydrotestosterone (DHT) and estradiol (E2) levels were measured. Total prostate volume (TPV) and transition zone volume (TZV) were measured by transrectal ultrasonography. Body mass index (BMI), DHT/TT and E2/TT were calculated. Correlations were determined using univariate and multivariate regression analyses.

Results. Apart from DHT, an age-dependent change of sex hormone levels were observed. On univariate analyses, age, BMI, serum DHT level and DHT/TT ratio, as well as serum E2 level and E2/TT ratio, but not serum TT and FT levels showed a significant association with prostate volume. On multivariate analysis, however, only serum DHT level and DHT/TT ratio remained significant. Logistic regression analysis showed that the odds ratios (95% confidence interval) of the second, third, and fourth quartiles of serum DHT levels for benign prostatic hyperplasia (defined as TPV ≥20 ml) risk were 2.06 (1.21–3.51), 2.66(1.56–4.53) and 7.15(4.0–12.6), respectively (p < 0.001).

Conclusions. Higher serum DHT level and DHT/TT ratio were associated with larger prostate volume and higher prevalence of BPH in Taiwanese men.

Keywords::

• Prostate
• testosterone
• dihydrotestosterone

Introduction

Benign prostatic hyperplasia (BPH) is the most common benign neoplasm in men. Although the pathogenesis of BPH is not completely elucidated, aging, sex hormones and growth factors are considered to play important roles in prostatic growth [1–5]. Several epidemiological studies have discussed the relationship between serum sex hormone concentrations and BPH, but the results have been inconsistent [6–19]. Testosterone is the main circulating androgen and has been the main focus of these studies. However, increasing data show that there is no association between serum testosterone concentrations and BPH, regardless of whether total testosterone (TT), free testosterone (FT), or bioavailable testosterone is measured [11,14–19 ]. Estrogens are also considered important regulators of prostate development. However, the relationship between serum estrogen levels and BPH risk remains uncertain [17].

Dihydrotestosterone (DHT) is a metabolite of testosterone and a more potent androgen. It is essential for the development of the male external genitalia during fetal development, and for the androgenic changes in skin and prostate in adults [1–5]. The conversion of DHT in the prostate is considered to be a major factor in BPH pathogenesis [1–5], as men castrated before puberty do not develop BPH [20], and those with an inherited deficiency of type 2 5α-reductase have only a vestigial prostate gland [21].

The correlation between serum DHT levels and prostate volume had been reported in patients receiving 5α-reductase inhibitors [22,23]. Parsons et al. also reported that in middle-aged individuals, higher midlife serum DHT levels were associated with increased BPH risk [18]. However, few large-scale epidemiological studies have been performed to identify the association between serum DHT levels and prostate volume, a surrogate of anatomic BPH, in healthy men. We conducted a cross-sectional study in a rather healthy Taiwanese cohort to evaluate the association between prostate volume and sex hormone levels, including serum DHT. We defined BPH as a prostate volume of 20 ml or greater as measured by transrectal ultrasonography. This definition is commonly used in epidemiological studies [24,25].

Materials and methods

Subjects

The subjects were men who participated in health examinations at National Taiwan University Hospital, and gave informed consent to participate in the study. Clinical data, including age, body mass index (BMI), tobacco and alcohol use, and International Prostate Symptom Score (IPSS), were obtained. Men with a malignancy or liver cirrhosis, men taking hormones, antiandrogen agents, antifungal agents or steroidal agents, or men who had been surgically or medically treated for BPH were excluded from the study.

Hormone measurements

Blood samples were collected before prostate evaluation, including digital rectal examination (DRE) and transrectal ultrasonography, and were taken between 8:00 and 11:00 AM to minimise the confounding effects of diurnal variation in hormone concentrations. Serum TT, estradiol (E2), FT and DHT levels were measured by radioimmunoassay (RIA) using DSL double antibody kits (Diagnostic Systems Laboratories Inc., Webster, TX). Intra-assay coefficient variabilities were 8.8% for TT, 7.1% for E2, 5.1% for FT and 7.8% for DHT.

Prostate evaluation

All subjects underwent DRE and transrectal ultrasonography (7.5-MHz, type 8551 probe and type 2001 Medical Ultrasound Scanner; B&K Medical, Denmark). Total prostate volume (TPV) and transition zone volume (TZV) were calculated by using the formula for the prostate ellipsoid: (π/6) × width × length × height. In this study, width was defined as the maximal transverse dimension, length as the maximal anteroposterior dimension and height as the maximal superior–inferior dimension. Subjects with abnormal DRE findings or elevated serum PSA levels were referred for prostate biopsy to exclude the possibility of prostate cancer. If biopsy results indicated prostate cancer, the subject was excluded from our study.

Statistical analysis

Data are expressed as mean (standard deviation; SD), median and quartile for continuous variables. In addition, we examined the ratios of serum E2 and DHT concentrations to serum TT concentrations to determine the relative amounts of these hormones. Univariate correlations between different parameters, including age, BMI, serum hormone levels and prostate volumes were estimated using Pearson's correlation analysis. A multivariate linear regression model including all parameters in which univariate analysis calculated p values < 0.1 was then constructed to assess the association between prostate volume and sex hormone levels. Serum sex hormone levels were categorised into quartiles based on distribution, and a logistic regression model was constructed to calculate odds ratio (95% confidence interval) for BPH (defined as TPV ≥20 ml) risk. Tests of linear trend across the quartiles were performed by using an ordinal variable corresponding to rank from the lowest category to the highest. P Values < 0.05 were considered to be significant.

Results

A total of 505 men (aged 40–79, mean age 58 years) were enrolled in our study. The clinical characteristics and endocrine parameters of the study cohort are summarised in Table I.

Table I.  Clinical and endocrine parameters.

	Mean (SD)	Median	Quartile
Age (years)	58.0 (9.5)	58	50, 66
BMI (kg/m^2)	25.1 (3.1)	25.1	23.1, 26.8
TPV (ml)	21.5 (8.9)	20.2	15.5, 25.8
TZV (ml)	8.4 (6.1)	6.7	4.5, 9.5
PSA (ng/ml)	1.3 (1.3)	0.9	0.5, 1.6
Total testosterone (ng/dl)	387.9 (127.2)	376	301, 450.5
Free testosterone (ng/dl)	12.8 (4.1)	12.4	9.8, 14.6
E2 (pg/ml)	39.3 (30.8)	33.9	14.5, 54.3
E2/total testosterone (%)	1.0 (0.8)	0.9	0.4, 1.4
DHT (pg/ml)	315.8 (288.5)	208.5	124.8, 393.0
DHT/total testosterone (%)	8.5 (7.6)	5.8	3.5, 10.6

BMI: body mass index; PSA: prostate specific antigen; DHT: dihydrotestosterone; TPV: total prostate volume; TZV: transition zone volume; E2, estradiol.

Figure 1 depicts the average serum sex hormone levels of these participants stratified into 5-year age groups. The mean serum TT, FT and E2 levels decreased significantly with age, whereas serum DHT levels had no significant association with age (Table II). In addition, age significantly associated with TPV (r = 0.29, p < 0.001) and TZV (r = 0.33, p < 0.001). A significant correlation between BMI and TPV (r = 0.16, p = 0.001) was also observed.

Figure 1.  Serum concentrations of sex hormones in different age groups. A: total testosterone; B: free testosterone; C: Estradiol; D: dihydrotestosterone. Boxes denote the median and interquartile ranges; bars indicate 5th–95th percentiles.

Table II.  Association between sex hormone levels and clinical parameters.

	TT	FT	E2	E2/TT	DHT	DHT/TT
Age	−0.23*	−0.40*	−0.14*	−0.06	0.06	0.15*
BMI	−0.15*	−0.11*	0.02	0.06	0.01	0.07
TPV	−0.004	−0.08	−0.11*	−0.13*	0.31*	0.33*
TZV	−0.14	−0.10	−0.11	−0.05	0.10	0.15*
TT	–	0.78*	0.36*	−0.05	0.23*	−0.13*
FT	–	–	0.49*	0.14*	0.06	−0.21*
E2	–	–	–	0.86*	−0.24*	−0.37*
E2/TT	–	–	–	–	−0.33*	−0.33*
DHT	–	–	–	–	–	0.89*
DHT/TT	–	–	–	–	–	–

BMI: body mass index; TPV: total prostate volume; TZV: transition zone volume; PSA: prostate specific antigen; TT: total testosterone; FT: free testosterone; E2: estradiol; DHT: dihydrotestosterone.

*p < 0.05.

Serum TT and FT levels were not significantly associated with TPV and TZV. In contrast, serum DHT levels and the DHT/TT ratio showed a significantly positive association with TPV and TZV. Conversely, serum E2 levels and the E2/TT ratio negatively associated with TPV (Table II). Multivariate linear regression analyses revealed that only serum DHT levels and the DHT/TT ratio significantly associated with TPV after adjusting for age, BMI and serum E2 level.

We then divided the patients into BPH (TPV ≥20 ml) and no BPH (TPV < 20 ml) groups. Patients with BPH were older and had higher BMI and TZV than did patients without BPH. There were no differences in serum TT and FT levels between the two groups. Serum DHT levels and DHT/TT ratios were significantly higher, and serum E2 level and E2/TT ratios significantly lower, in the BPH group (Table III).

Table III.  Comparisons between patients with BPH (TPV ≥20 ml) and without BPH (TPV < 20 ml).

	BPH (n = 258) TPV ≥20 ml	No BPH (n = 247) TPV <20 ml	p value
Age (years)	59.8 (9.0)	56.1 (9.6)	<0.001
BMI (kg/m^2)	25.5 (3.2)	24.8 (2.8)	0.020
TZV (ml)	9.6 (6.6)	4.9 (1.9)	<0.001
Total testosterone (ng/dl)	392.8 (136.9)	382.9 (116.4)	0.540
Free testosterone (ng/dl)	12.5 (3.9)	13.1 (4.3)	0.345
E2 (pg/ml)	35.3 (30.8)	43.3 (30.3)	<0.001
E2/total testosterone (%)	0.9 (0.8)	1.2 (0.8)	<0.001
DHT (pg/ml)	398.5 (327.0)	229.5 (210.0)	<0.001
DHT/total testosterone (%)	10.7 (8.7)	6.2 (5.4)	<0.001

BMI: body mass index; E2, estradiol; TPV: total prostate volume; TZV: transition zone volume; DHT: dihydrotestosterone.

Logistic regression analysis showed that higher serum DHT levels and DHT/TT ratio were associated with a significantly increased risk of BPH (Table IV). The odds ratios (95% confidence interval) for BPH risk in the second, third and fourth quartiles of serum DHT levels relative to the lowest quartile were 2.06 (1.21–3.51), 2.66 (1.56–4.53) and 7.15 (4.04-12.6), respectively (p trend <0.001).

Table IV.  Logistic regression for BPH after adjusting age and BMI.

	Quartile	OR (95% CI)	p value
DHT	1 (referent)	1.00
	2	2.06 (1.21–3.51)	0.008
	3	2.66 (1.56–4.53)	<0.001
	4	7.15 (4.04–12.6)	<0.001
	p trend		<0.001
DHT/TT	1 (referent)	1.00
	2	1.96 (1.15–3.32)	0.013
	3	3.11 (1.82–5.33)	<0.001
	4	5.32 (3.05–9.29)	<0.001
	p trend		<0.001

DHT: dihydrotestosterone; TT: total testosterone; OR: odds ratio; CI: confidence interval.

Discussion

Although the role of sex hormones in causing BPH remains elusive, these hormones undoubtedly play at least a permissive role in prostate growth [1–5]. However, the gradual decline in serum testosterone levels in older men beginning from middle age coincides paradoxically with the period of progressive prostate growth. Serum TT and FT levels had no significant association with prostate volume in our study, a finding that is consistent with prior reports of inverse or null associations between serum testosterone levels and BPH [11,14–18]. However, this lack of association does not nullify the androgen hypothesis; rather, it underscores the importance of better understanding the mechanism of androgen action within the prostate, including the relationship between tissue and serum androgen levels.

In contrast to the significant age-dependent decrease in serum TT and FT levels, serum DHT levels remained constant throughout the age groups in our study. In males of most species, DHT is present in the plasma at a concentration equal to approximately 10% of testosterone concentration [26]. However, serum DHT levels reflect not only the activity of 5α-reductase type 2 within the prostate, but also that of 5α-reductase type 1 in the skin and liver [1–5]. Because serum DHT levels do not necessarily correlate to prostate tissue DHT concentrations [27], some reports examined serum DHT metabolites, such as 17β-diol glucuronide or androstanediol glucuronide (AAG) as a surrogate for measuring DHT activity, whereas other reports adopted serum DHT or DHT metabolites to TT ratio as an indirect indicator of 5α-reductase activity [28,29]. Norman et al. [30] observed a strong association between changes in serum and intraprostatic DHT levels, but not between changes in serum AAG and intraprostatic DHT levels. The group suggested that serum DHT levels might better reflect short-term changes in 5α-reductase activity, and that the larger pool of serum AAG might be less affected by short-term variations in enzyme activity.

To our knowledge, our study is the first to report a positive association between serum DHT levels (and DHT/TT ratio) and prostate volume in a healthy cohort. The significant association found in our study implies that men with higher serum DHT levels or DHT/TT ratios may have higher 5α-reductase activity and higher likelihood of developing BPH. Although previous cross-sectional studies regarding the association between DHT and prostate volume were inconsistent [9,13,14,31], our results are supported by two large prospective studies [17,18] and clinical experience with 5α-reductase inhibitors [22,23]. Parsons et al. [18] reported that higher midlife serum DHT levels were associated with increased BPH risk, whereas higher TT/DHT ratios were associated with reduced BPH risk. Kristal et al. [17] also reported that serum TT:17β-diol glucuronide ratio was associated with reduced BPH risk.

5α-Reductase activity may vary among different races. For example, the DHT/TT ratio had been reported to be highest in African-Americans, intermediate in whites and lowest in Asian-Americans [31], which corresponds to the respective BPH incidence rates in these groups. Litman et al. [32] reported that there are no ethnic differences in serum TT levels between African-American, Latino and white men. However, a marginal ethnic difference in DHT/TT ratio was also found.

The role of estrogen in BPH pathogenesis is more complex. Some investigators reported an inverse association between E2 and BPH [10,17] whereas others have observed a positive association [6–8,11] or no association [18]. We found a significant negative association between TPV and serum E2 levels (and E2/TT ratios) after adjusting for age and BMI, but the associations attenuated after adjusting for serum DHT levels. DHT may play a major role in prostate growth, and the actual role of estrogens remains elusive.

The strength of our study was the use of an objective ultrasound measure of prostate volume, which overcomes a major limitation of previous studies using clinical diagnosis of BPH. In addition, both age and BMI were adjusted for regression analyses, because higher amounts of body fat enhance peripheral conversion of testosterone to estrogen. Certain potential limitations should also be noted. The cross-sectional study design precludes the ability to make causal inferences from the observed association. Also, serum hormones were only measured once. The single measurement provides an imperfect estimate of a subject's usual hormonal status, and can be influenced both by individual errors and analytical errors.

In conclusion, our study in relatively young and healthy Taiwanese men showed that higher serum DHT levels and DHT/TT ratios were associated with larger prostate volume and higher BPH prevalence. In addition, we also know that 5α-reductase activity may vary among different races and individuals. This significant association may imply that men with higher serum DHT levels or DHT/TT ratios have higher 5α-reductase activity, which increases the likelihood of developing BPH. Since the aetiologies of BPH remain elusive, measuring serum DHT level may provide auxiliary information for patient's counselling on BPH. It is of interest to investigate whether current serum DHT level may provide as a useful predictor for future BPH development and progression among middle-aged men. Also, it will be of interest to see whether serum DHT level can predict the treatment response among patients who are receiving 5α-reductase inhibitor therapy.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.