Lower urinary tract symptoms (LUTS) are a set of bothersome voiding symptoms, including increased urinary frequency, urgency, nocturia, slow stream, postmicturition dribble, and are traditionally attributed to benign prostatic hypertrophy (BPH). Epidemiological studies have suggested that age is among the principal unmodifiable risk factor of LUTS [1]. Metabolic syndrome (Met S) is a complex pathology manifested through a host of interrelated factors, such as increased waist circumference (WC), elevated insulin resistance (IR), hyperglycemia, hypertension, visceral adiposity, dyslipidemia, endothelial dysfunction, and atherosclerotic disease concomitant with low-grade inflammation, as indicated by elevated non-specific inflammatory markers. The Met S components often exist concomitantly with a reduction in circulating serum testosterone (T) levels [2345678910111213]. Furthermore, Met S is thought to be associated with nephrolithiasis, BPH, LUTS, erectile dysfunction (ED), and infertility [14]. Met S and its related comorbidities are often associated with development and/or progression of BPH-LUTS [1].

Preclinical and clinical studies suggest that obesity, dyslipidemia, diabetes, components of Met S are critical determinants and are potentially modifiable risk factors in BPH-LUTS (Fig. 1). Testosterone deficiency (TD) is a predictor and contributor to Met S [23411]. Met S is associated with increased inflammatory state and increased sympathetic activity together with pelvic ischemia [14]. These pathophysiological states contribute to the worsening of LUTS. Furthermore, TD contributes to increased adrenergic activity and reduced nitric oxide synthase activity concomitant with increased endothelin and Rho kinase activity, further contributing to the worsening of LUTS.

Central obesity and WC, components of Met S, have been associated with increased risk for BPH and LUTS [1415161718192021222324252627282930]. However, central obesity may adversely impact LUTS independent of the prostate growth, as demonstrated by changes in severity of LUTS in men after prostatectomy [31323334]. An emerging body of evidence suggests that a relationship exists between LUTS severity and Met S [3233353637]. As reported by Kupelian et al [38], Met S was significantly associated with an elevated American Urological Association-symptom index score (multivariate odds ratio [OR]=1.68). After adjusting for age, men younger than 60 years with Met S were more likely to report intermittency, incomplete emptying, and nocturia [38]. We should point out that some studies [21] suggested that the presence of Met S was not significantly associated with moderate to severe LUTS (OR=1.13, p=0.53) and only altered serum triglycerides, an important component of Met S, and diabetes were associated with this risk. Yet, a positive association was noted between number of Met S components and LUTS/bladder outlet obstruction in this study [21]. It is possible that endocrine abnormalities attributed to central obesity and hypertriglyceridemia may play an important role in development or progression of the pathophysiology of lower urinary tract dysfunction [3839]. In particular, triglyceride and cholesterol levels seem to have a detrimental effect on prostatic function and inflammation, which may be associated with development and progression of LUTS/BPH [37]. Moreover, a meta-analysis, including 8 studies and 5,403 patients, demonstrated that Met S is associated with increased prostate size, supporting a potential role for metabolic dysfunction in progression of BPH, thus contributing to the pathophysiology of LUTS. It is likely that obesity, dyslipidemia and age represent significant risk factors in worsening LUTS [32]. Furthermore, an association between Met S and pelvic floor dysfunction exists independent of disease of the prostate [40].

As shown in Table 1, a considerable body of literature exists demonstrating a bidirectional link between TD and Met S [23441424344454647484950515253545556575859606162]. Several systematic reviews, crosssectional studies, and clinical trials have reported low levels of total T (TT) and sex-hormone binding globulin (SHBG) in men with Met S, irrespective of age [11446364]. The relationship between low TT and SHBG may be attributed to hyperinsulinemia in men with Met S. In addition, it was shown that serum estradiol (E₂) levels are higher in men with Met S and are positively correlated with the number of Met S components [65]. Reduced TT levels predict the development of the Met S in men with normal body mass index (BMI) [41435666]. Additional evidence that low TT levels increase the risk of Met S is derived from experience with androgen deprivation therapy for prostate cancer [676869].

Laaksonen et al [41] evaluated 702 men who initially were without Met S or diabetes mellitus. After 11 years of follow-up, there were 147 incident cases of Met S (20.9%) and 57 incident cases of diabetes (8.1%). After adjustment for age, men with TT, free testosterone (FT) and SHBG levels in the lower T quartile had a several fold increased risk of developing Met S (OR=2.3, 95% confidence imterval [CI], 1.5–3.4; OR=1.7, 95% CI, 1.2–2.5; and OR=2.8, 95% CI, 1.9–4.1, respectively) and diabetes (OR=2.3, 95% CI, 1.3–4.1; OR=1.7, 95% CI, 0.9–3.0; and OR=4.3, 95% CI, 2.4–7.7, respectively). Furthermore, adjustment for potential confounders, such as cardiovascular disease, smoking, alcohol intake, and socioeconomic status did not alter the associations. The authors concluded that low TT and SHBG levels independently predict development of Met S and type 2 diabetes in middle-aged men. In addition, it was suggested that TD is an early marker for disturbances in both insulin and glucose metabolism, which may progress to Met S and diabetes. This may also contribute independently to the pathogenesis of such diseases. In a subsequent study, Laaksonen et al [42] reported that men with Met S at baseline had a 2.6-fold increased risk of developing TD (T levels<11 nM) at the 11-year follow-up independent of age, smoking, and other potential confounders. In addition, adjustment for BMI (OR=2.0, 95% CI, 1.1–3.8) or baseline TT levels (OR=1.9, 95% CI, 1.0–3.4) attenuated the association. The association of Met S with TD (FT<225 pmol/L) was similar. The adjusted decrease in T levels during the 11-year follow-up was also greater in men with than without Met S. Antonio et al [56] reported that in 1,651 men without Met S at baseline, 289 men developed incident Met S during the follow-up. Men with lower baseline TT levels were at higher risk for developing Met S (OR=1.72, p<0.001), even after adjustment for SHBG (OR=1.43, p<0.001), BMI (OR=1.44, p<0.001), or homeostasis model assessment of insulin resistance (HOMA-IR) (OR=1.64, p<0.001). E₂ was not associated with development of Met S (OR=1.04, p<0.56). However, a lower E₂/T ratio was associated with a lower risk of incident Met S (OR=0.38, p<0.001), even after adjustment for SHBG (OR=0.48, p<0.001), BMI (OR=0.60, p<0.001) or HOMA-IR (OR=0.41, p<0.001). It was concluded that lower T levels, but not E₂, are linked with an increased risk of developing Met S, independent of SHBG, BMI, or IR. A lower E₂/T ratio may be protective against developing Met S [56]. The authors suggested that low T but not E₂ levels in men may be regarded as a biomarker or risk predictor for Met S, independently of SHBG, IR, and body composition. The importance of aromatase activity in Met S requires further investigation. These findings may have implications for the assessment of cardiometabolic risks in older and obese men [60]. The aforementioned studies strongly suggest that low TT levels independently predict Met S and the number of Met S components are inversely correlated with TT levels. Considering the Met S components individually, WC seems to be the strongest correlated factor with lower TT levels as demonstrated by multivariate analysis [11]. Therefore, it is reasonable to suggest that TD is an independent predictor of the Met S.

T therapy has been shown to improve hyperglycemia and IR and ameliorate some of the Met S components (Table 3) [116117118119120121122123124125126127128129]. However, conflicting evidence exists suggesting changes in weight loss and WC are not maintained [63]. Mårin et al [130] reported that T therapy improved insulin sensitivity in obese men and reduced central adiposity. Recently it has also been shown that T therapy improves insulin sensitivity in the HOMA-IR model. A significant reduction in IR in men with TD and type 2 diabetes was reported in men treated with intramuscular T therapy over a 3-month period [131]. Although this study reported reduction in glycated hemoglobin and fasting blood glucose, the effect of T therapy on these factors remains unclear [123131]. In one multicenter, double-blind, placebo-controlled trial, T treatment reduced hemoglobin A1c (HbA1c) and WC [132]. Heufelder et al [117] also demonstrated this in 32 hypogonadal (TT<12 nmol/L) men with Met S and newly diagnosed type 2 diabetes randomized to 52 weeks of treatment with diet and exercise with or without transdermal T (50 mg/d). Addition of T treatment to supervised diet and exercise resulted in greater therapeutic improvement of glycemic control and reversed the Met S in 81% vs. 31% of controls after 52 weeks. Hoyos et al [133] investigated the effects of T therapy in obese men with severe obstructive sleep apnoea in an 18-week randomized, placebo-controlled trial. Sixty-seven men were randomized to intramuscular T therapy or placebo. T therapy improved insulin sensitivity, reduced liver fat, and increased muscle mass, as well as decreased arterial stiffness, but did not differentially reduce overall weight or the Met S and this may be attributed to the short treatment duration.

Cai et al [134] have identified only five randomized clinical trials with a total of 351 subjects who were treated with T for an average of 6.5 months. A meta analysis of these clinical trials to determine the metabolic effects of T therapy in hypogonadal men with type 2 diabetes found that fasting plasma glucose levels decreased 1.1 mmol/L (95% CI, −1.88 to −0.31), fasting serum insulin levels decreased by 2.73 mIU/L (95% CI, −3.62 to −1.84), HbA1c decreased by 0.87% (95% CI, −1.32% to −0.42%). The authors concluded that T therapy can improve glycemic control and decrease triglyceride levels in men with TD and type 2 diabetes.

Interestingly, other trials have failed to show improvements in glucose metabolism. Gianatti et al [135] conducted a randomized, double-blind, parallel, placebo-controlled trial in 88 men with type 2 diabetes, aged 35 to 70 years with an HbA1c ≤8.5% (69 mmol/mol), and a TT level of ≤12.0 nmol/L (346 ng/dL). Participants were randomly assigned to 40 weeks of intramuscular T therapy (n=45) or matching placebo (n=43). T therapy did not improve IR despite a decrease in fat mass and an increase in lean mass. T therapy reduced subcutaneous but not visceral abdominal adipose tissue. T therapy did not improve glucose metabolism or visceral adiposity in obese men with moderately controlled type 2 diabetes and modest reductions in circulating T levels typical for men with type 2 diabetes.

Haider et al [136] reported data from a registry study on T therapy in men with obesity and diabetes. A total of 156 obese, diabetic men with TD, aged 61.17±6.18 years, fulfilled selection criteria. T treatment for up to 6 years decreased WC by 11.6 cm, weight declined by 17.5 kg (15.04%), fasting glucose declined from 7.06±1.74 to 5.59±0.94 mmol/L (127±31 to 101±17 mg/dL, p<0.0001) and HbA1c decreased from 8.08% to 6.14%, with a mean change of 1.93%. Systolic and diastolic blood pressure, lipid profiles including total cholesterol: HDL ratio, C-reactive protein, and liver enzymes all improved (p<0.0001). While this is not a randomized controlled clinical trial, it does suggest the potential benefits of T treatment on metabolic parameters when weight loss is achieved, and it suggests that long term treatment with T may contribute to weight loss.

It is widely recognized that T therapy reduces fat mass and increases lean body mass; however, until recently, most reports have not been associated with much weight loss [137]. Changes in body composition and weight loss are considered potential mechanisms by which T treatment improves insulin sensitivity and glucose control in patients with diabetes. Effects on inflammatory cytokines [138] and changes in oxidative metabolism [139] also have been reported to improve glucose metabolism. In a HFD induced model of Met S, Maneschi et al [8] demonstrated that Met S is associated with visceral adipose tissue ischemia, which contributes to altered cytokine production and endocrine signaling. They demonstrated that the insulin-regulated GLUT4 transporter expression was reduced by a HFD, but that this was reversed with T therapy administration. These findings suggest that adipose tissue signaling is modulated by T and that T therapy could potentially contribute to normalization of adipose function in subject suffering from Met S.

It has been shown by a number of studies that T therapy improves LUTS (Table 4) [73126128140141142143144145]. According to the study by Favilla et al [78] in men with BPH a significant association exists between severity of LUTS and reduced serum TT level, but not other serum sex hormones. Importantly, this study demonstrated that serum T levels are not associated with prostate size and volume. In fact, it has been shown that there exists a negative relationship of FT and BT with IPSS on multivariate analysis [24]. This negative relationship provides a framework with which to approach the evidence regarding T Therapy in men with LUTS.

The way in which T is hypothesized to improve LUTS includes regulation of the expression of alpha1-adrenergic receptors, phosphodiesterase type 5 activity, Rho-kinase activation, endothelin activity, and neural nitric oxide synthase [146147]. As shown in an animal model, T therapy also improves LUTS and bladder function parameters, in part by increasing bladder capacity and compliance and by decreasing detrusor pressure at maximal flow [106]. In that animal model, the authors found that orchiectomy decreased the mean bladder capacity by 38.9% while T therapy actually increased bladder capacity by an average of 46.6%. Celayir [107] measured the baseline urodynamic and T levels in control and orchiectomized rabbits treated with or without T. Bladder capacity and compliance were increased on days 5 and 10 with T treatment but decreased thereafter and returned to the baseline levels on day 30, suggesting that T modulates bladder capacity and compliance.

Aside from storage factors, T therapy is also observed to decrease DSM cell excitability by direct activation of large conductance voltage and calcium activated K+ channels through a non-genomic mechanism [111]. Additionally, T therapy can induced bladder neck smooth muscle relaxation and inhibit contractility in the DSM [109110]. Such findings suggest that there are underlying T-mediated physiological and anatomical mechanisms in bladder function.

It has long been believed that the most common etiology for LUTS in older men is BPH [108]. An inverse associations between serum T and LUTS was reported by Trifiro et al [88]. T therapy was reported to improve lower urinary function by increasing bladder capacity and compliance and by decreasing detrusor pressure at maximal flow in men TD [140141].

Crawford et al [148] found that endogenous T levels did not correlate with LUTS or prostate size and these findings support the saturation theory [149]. This is consistent with data reported by Schatzl et al [85], in which T levels were shown to have no impact on LUTS status or prostate volume. On the other hand, preliminary evidence indicates that men with LUTS benefit from T treatment [116] and pilot studies have also shown that T therapy has a positive effect on LUTS men with TD [140]. In a recent review Baas and Köhler [150] suggested that T may be beneficial for BPH/LUTS, likely via increased expression and activity of nitric oxide synthase concomitant with smooth muscle relaxation. Concern remains that T therapy may worsen LUTS secondary to BPH, although few major adverse events have been reported with T treatment. We examined the effects of long-term T therapy in men with TD on urinary and erectile function and compared the findings with those of men in the same registry who remained untreated. More importantly we performed propensity matching analysis [151] to adjust for a number of variables to ensure that confounding factors such as age and obesity were considered. This was done to assess changes in urinary and sexual function as well as in quality of life (QoL). The currently reported data and those reported previously [73151] were derived from the same registry. We report that T therapy produced significant improvements in LUTS as assessed subjectively by IPSS and objectively by the significant reduction in post-voiding residual volume (PVR). Improvements in IPSS and PVR were progressive and sustained during the entire follow-up of approximately 8 years. Further analysis of IPSS showed a clear difference between the treated and untreated groups. Others also reported that T therapy significantly increased mean maximal bladder capacity and compliance (p=0.007 and p=0.032, respectively) while mean detrusor pressure at maximum urine flow significantly decreased from before to after treatment (p=0.017) [141].

Amano et al [142] treated 41 men with TD (mean baseline FT of 6.8±1.5 pg/mL) using a 1% T ointment. Prior to treatment total IPSS ranged from 0 to 25 (mean score, 8.5±6.5), suggesting mild to moderate LUTS symptoms. After three months of treatment, the mean FT increased markedly (6.82±1.51 to 8.24±2.84 pg/mL) and total IPSS significantly decreased (mean, 6.0±5.4, p=0.0086), Most importantly, all sub-scores of IPSS, including post-voiding, storage, and voiding symptoms and QoL check to see if abbreviated all improved with T therapy [142].

Debruyne et al [152] reported on 999 men with clinically diagnosed TD and 750 men (75%) initiated T therapy. There were no differences in PSA levels, total IPSS, or the IPSS obstructive sub-scale score in response to T therapy. Lower IPSS irritative sub-scale scores were reported in T treated compared to untreated men. Moderate to severe LUTS (IPSS≥8) were reported at baseline by 38.3% of untreated men and 41.1% in treated men. Adjusted and unadjusted longitudinal models showed modest, positive effects of T therapy on LUTS, with 7.1% lower total IPSS over time (p=0.004) and 7.9% reduction of irritative scores (p<0.001) in T treated compared with untreated men, but no change in IPSS obstructive scores (p=0.33). Shigehara et al [143] found that at 6 months, T therapy with 250 mg of T enanthate every 4 weeks was associated with decreased nocturia, improved sleep conditions, and improved QoL, which in and of itself has been shown to improve endogenous T production [153]. T treatment has also been shown to improve voiding function, urinary symptoms and changes in bladder wall thickness observed with androgen deficiency [140141145].

Several studies have suggested a relationship between LUTS and Met S, and it is accepted that severe LUTS are more likely to develop in men with the Met S [96154155]. In a recent study of the epidemiological relationship between Met S and LUTS it was hypothesized that Met S was associated with over activity of the autonomic nervous system and that IR, a key element of Met S, might be responsible [154155]. Furthermore, the presence of Met S might mediate intraprostatic inflammation because of its association with an increased serum high sensitivity C-reactive protein concentration, which would link Met S to symptomatic BPH [96156].

Liu et al [157] investigated the effects of T therapy on LUTS in 632 diabetic men and reported that men with the lowest T levels (2.21±0.51 ng/dL) had the highest prevalence of nocturia, and it was more likely to be severe (≥3 times/night). Francomano et al [126] examined the effects of T therapy for 5 years on LUTS in 20 obese men with TD and Met S and compared the findings in 20 matched subjects who remained untreated. After 5 years of T therapy no significant changes were reported in IPSS, PVR, or Q_max. The authors suggested that the improvement in the components of the Met S may have played some role in maintaining unaltered urinary function. Furthermore, in a propensity matched subgroup analysis of 164 men with low T, it was shown that long term T therapy of as much as 10 years contributed to improved IPSS and Aging males' symptom scores, as well as decreased post-void residual volume [73].

Kohn et al [158] reported that marked differences exist in LUTS between obese and non-obese patients, including differences in voiding and storage symptoms. The increased obesity associated with TD and worsening LUTS may have a common pathophysiological mechanism. Obesity increases aromatase activity, leading to a state of TD, while in the meantime the prostate continues to enlarge, worsening LUTS [126]. Since T therapy results in a sustained and marked decrease in body weight, WC, and BMI [151], weight loss associated with T treatment may contribute to the observed improvement in LUTS. Our findings are also supported by a number of human and animal studies demonstrating a strong independent association between the components of Met S and BPH/LUTS as discussed above [153].