Prevalence of low testosterone and its relationship to body mass index in older men with lower urinary tract symptoms associated with benign prostatic hyperplasia

Abstract

Purpose: We examined the prevalence of low testosterone (LT) and its relationship with body mass index (BMI) in men with lower urinary tract symptoms (LUTS) associated with benign prostatic hyperplasia (BPH), who were enrolled in a clinical trial of drug therapy, the Medical Therapy of Prostatic Symptoms (MTOPS) Study.

Materials and methods: MTOPS enrolled 3047 men, and of these, 1896 had total testosterone (TT) measured at baseline. LT was defined as a single measurement of TT of <300 ng/dL.

Results: The overall prevalence of LT was 25.7%. Prevalence increased with increasing BMI; 14.7% among men who were normal weight (BMI <25 kg/m²) and 24.2% and 39.3% among overweight (BMI 25 to <30 kg/m²), and obese (baseline BMI ≥30 kg/m²) men, respectively.

Conclusions: LT was observed in about one in four MTOPS study participants with baseline TT measurements. The prevalence of LT increased markedly with increasing BMI. Our findings suggest a high prevalence of LT in obese men with LUTS/BPH. Physicians should be alert to the possibility of symptoms of hypogonadism in this population.

• Benign prostatic hyperplasia
• lower urinary tract symptoms
• Medical Therapy of Prostatic Symptoms trial
• testosterone deficiency

Introduction

Hypogonadism in men is a clinical syndrome that results from failure of the testes to produce physiological levels of testosterone (androgen deficiency) and a normal number of spermatozoa due to disruption of one or more levels of the hypothalamic-pituitary-testicular (HPT) axis. Symptoms of hypogonadism are associated with low levels of serum total testosterone (TT) (<300 ng/dL) [1]. It is well established that increased body mass index (BMI) is a risk factor for both benign prostatic hyperplasia (BPH) and low testosterone levels (LT) in aging men [2–4]. However, to date, only two small studies have examined the prevalence of low testosterone levels in BPH patients. Each of these studies found that approximately one in five men had LT; with an increased prevalence in obese men [5,6]. We examined the prevalence of LT (here defined as TT <300 ng/dL) and the relationship between LT and BMI in a large subset of men enrolled in the Medical Therapy of Prostatic Symptoms (MTOPS, NCT00021814) trial, a clinical study to determine the effects of long-term medical therapy on clinical progression of BPH in men with lower urinary tract symptoms (LUTS) secondary to BPH.

Methods

The MTOPS trial enrolled men at least 50 years of age with an American Urological Association Symptom Index score of 8–30 (8–35 in the pilot phase), a maximum urinary flow rate (Qmax) of 4–15 mL/s and a voided volume of at least 125 mL. A total of 3047 men were enrolled, including 116 from the pilot study. We studied baseline TT data from 1198 of these men who participated in a sub-study that included measurement of hormones and a prostate biopsy at baseline and at pre-determined time points during follow-up [7]. We also examined baseline TT data from an additional 698 MTOPS patients (not included in the sub-study described earlier) who participated in the full-phase study and were independently evaluated for TT at baseline by trial physicians. Thus, the overall MTOPS TT analysis population included a total of 1896 men.

We defined men with LT as having a single measurement of TT <300 ng/dL. Men were grouped according to the following categories of BMI: <25 kg/m² (normal weight), ≥25 to <30 kg/m² (overweight) or ≥30 kg/m² (obese) and the percentage of patients with LT in each BMI group was determined.

Serum testosterone was measured using a radioimmunoassay kit (Testosterone Coat-A-Count) manufactured at that time by DPC (Los Angeles, CA, USA) now available from Cruinn Diagnostics Limited (Dublin, Ireland). BMI was calculated as (weight (lb)/[height (in)]²) × 703.

Results

Characteristics of study participants

Selected baseline characteristics of the 1896 men who had TT measured at baseline are shown in Table 1. Overall, they had a mean baseline TT of 397.4 ng/dL, a mean BMI of 27.7 kg/m² and a mean age of 62.6 years; 25.8% were normal body weight, 48.1% were overweight and 26.1% were obese. The mean age and BMI of the subgroup analyzed here was similar to the remaining men enrolled in the trial who did not have TT measurements (data not shown). Men with LT (baseline TT<300 ng/dL) had a mean baseline TT of 235.6 ng/dL, mean BMI of 29.6 kg/m² and mean age of 62.7 years. Men with normal baseline TT (baseline TT ≥ 300 ng/dL) had a mean baseline TT of 453.4 ng/dL, mean BMI of 27.1 kg/m² and mean age of 62.5 years. Men in each of the three BMI categories examined in this analysis (normal weight, overweight and obese) were similar with respect to mean age.

Table 1. Patient population characteristics.

	N	Mean age ± SD	Age range	Mean TT ± SD	Mean BMI ± SD
Total population	1896	62.6 ± 7.3	50–86	397.4 ± 147.5	27.7 ± 4.2
BMI <25 kg/m^2	489	63.9 ± 8.1	50–86	448.4 ± 154.6	23.1 ± 1.4
BMI ≥25, <30 kg/m^2	913	62.2 ± 7.2	50–83	395.5 ± 141.6	27.3 ± 1.4
BMI ≥30 kg/m^2	494	61.9 ± 6.5	50–80	350.6 ± 134.5	33.2 ± 3.2

BMI, body mass index; SD, standard deviation; TT, serum total testosterone.

Prevalence of low testosterone levels

Three-fourth (75%) of MTOPS participants we studied were overweight or obese. Overall, one-fourth (25.7%) had LT. The prevalence of LT increased with increasing BMI; it was 14.7% in normal weight men, 24.2% in overweight men and 39.3% in those who were obese (Figure 1). Among the 487 men with LT, most (85%) were overweight or obese.

Figure 1. Prevalence of low testosterone (baseline serum total testosterone <300 ng/dL), by baseline BMI.

Discussion

Among aging men with LUTS secondary to BPH, about one in four were found to have LT. The prevalence of LT we observed in a group of men selected for LUTS/BPH is within the range of prevalence reported for the general male population >30 years of age [8–11], though at least one study found the prevalence of LT in the general population to be substantially lower [12]. We observed the highest prevalence of LT in obese LUTS patients (39.3%), with decreasing prevalence in overweight (24.2%) and normal weight (14.7%) patients. The increased prevalence of LT in obese patients was similar to that observed in the Proscar Long-Term Efficacy and Safety Study, in which 36.1% of 61 obese BPH patients had LT at baseline [6]. The well-established relationship between LUTS/BPH and increased BMI [2–4] was reconfirmed in our study as the majority (75%) of men were overweight or obese.

The results of this analysis are consistent with previous work in which LT was observed to be associated with increased BMI [6,13–17], or with waist circumference [18]. There are a number of mechanisms that may explain the inverse relationship between higher BMI and testosterone levels in men. Adipose tissue is the main peripheral source of aromatase, which catalyzes the irreversible conversion of T to estradiol. The increased adiposity in obesity may be associated with increased aromatase activity, which could lead to an increase in the conversion of T to estradiol [19]. Obese men have been shown to have elevated serum estradiol levels [20]. Increased estradiol levels may, in turn, cause pituitary suppression [19], which may explain, at least in part, the reduction in gonadotropin release reported in obese men [19,20]. Obesity has also been shown to be associated with reduced sex hormone-binding globulin levels, which could lead to a further reduction in serum testosterone levels [20,21].

In light of the role testosterone plays in maintaining male sexual function [22,23], together with its roles in regulation of bone density, muscle mass and function, fat mass and cardiovascular fitness [24,25], physicians should be aware of the relatively high prevalence (39.3%) of LT we observed in obese men with LUTS secondary to BPH. Consideration might also be given to the possibility that the symptoms associated with LT could exacerbate the significant deterioration in quality of life associated with LUTS [26,27]. Practice guidelines [1] for men with significant symptoms of androgen deficiency and LT include testosterone replacement therapy. However, physicians managing LUTS patients with LT are presented with a possible dilemma. It has been reported that in some patients testosterone therapy can lead to significant increases in both prostate volume and prostate-specific antigen (PSA) [28–32]. As both prostate volume and serum PSA levels positively correlate with acute urinary retention and BPH-related surgery [33,34] there is concern that a stimulatory effect of testosterone therapy on the prostate could worsen urinary symptoms and offset the beneficial effects of medical therapies for LUTS/BPH (alpha-blockers and 5α-reductase inhibitors). Moreover, clinically significant elevations in serum PSA reported with some testosterone therapies [28] may complicate the interpretation of PSA levels in men using 5α-reductase inhibitors. However, in studies that evaluated the effect of testosterone replacement therapy on urinary symptoms, no changes were noted [29–31]. Furthermore, a meta-analysis of testosterone supplementation trials [35] concluded that therapy does not increase either prostate size or PSA levels, or worsen urinary function.

Our findings that men with LUTS secondary to BPH exhibit a relatively normal range of serum testosterone, rather than being skewed towards higher levels, are consistent with the observations made in the meta-analysis [35]. However, as there may be variation among individuals in response to testosterone replacement therapy, particularly with regard to prostate volume changes [32], physicians administering testosterone replacement to patients with LT should consider routinely monitoring prostate and urinary status during therapy [31].

Limitations of this analysis include the non-random nature of the study population, the cross-sectional study design, use of a single measure of testosterone, the limited racial/ethnic composition of the study group (mostly Caucasian), likely variability in the time of day a blood sample was collected, and absence of information on clinical status (symptoms related to low testosterone).

Conclusions

In conclusion, the present analysis of baseline TT data from 1896 men participating in the MTOPS trial demonstrated that about one in four LUTS patients, and greater than one in three obese LUTS patients, had LT. In light of these findings, physicians may want to consider examining men with LUTS/BPH for concomitant androgen deficiency.

Declaration of interest

S.A.K. has received compensation from Merck & Co., Inc. A.G.M. and E.A.O. are employees of Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Whitehouse Station, NJ, USA, and may hold stock/stock options in the company. J.W.K. and J.Y.L. have no disclosures to report. MTOPS was supported by the following cooperative agreements from the National Institute of Diabetes, Digestive and Kidney Diseases: U01 DK49977, U01 DK46416, U01 DK41418, U01 DK46429, U01 DK46431, U01 DK46437, U01 DK46468, U01 DK46472, U01 DK49880, U01 DK 49912, U01 DK49921, U01 DK49951, U01 DK49954, U01 DK49960, U01 DK49961, U01 DK49963, U01 DK49964, U01 DK49971, U01 DK49980, as well as by the National Center for Minority Health and Health Disparities, NIH. Financial support and drug products for MTOPS were also provided by Merck and Pfizer.

Acknowledgements

The authors wish to acknowledge the contributions of the many investigators, study coordinators and patients who contributed to the MTOPS trial. Editorial assistance was provided by Jennifer Rotonda, Ph.D. of Merck & Co., Inc., Whitehouse Station, NJ, USA.