Metabolic syndrome and prostate abnormalities in male subjects of infertile couples.

No previous study has evaluated systematically the relationship between metabolic syndrome (MetS) and prostate-related symptoms and signs in young infertile men. We studied 171 (36.5 ± 8.3-years-old) males of infertile couples. MetS was defined based on the National Cholesterol Education Program Third Adult Treatment Panel. All men underwent hormonal (including total testosterone (TT) and insulin), seminal (including interleukin-8 (IL-8), seminal plasma IL-8 (sIL-8)), scrotal and transrectal ultrasound evaluations. Because we have previously assessed correlations between MetS and scrotal parameters in a larger cohort of infertile men, here, we focused on transrectal features. Prostate-related symptoms were assessed using the National Institutes of Health Chronic Prostatitis Symptom Index (NIH-CPSI) and the International Prostate Symptom Score (IPSS). Twenty-two subjects fulfilled MetS criteria. In an age-adjusted logistic ordinal model, insulin levels increased as a function of MetS components (Wald = 29.5, P < 0.0001) and showed an inverse correlation with TT (adjusted r = -0.359, P< 0.0001). No association between MetS and NIH-CPSI or IPSS scores was observed. In an age-, TT-, insulin-adjusted logistic ordinal model, an increase in number of MetS components correlated negatively with normal sperm morphology (Wald = 5.59, P< 0.02) and positively with sIL-8 levels (Wald = 4.32, P < 0.05), which is a marker of prostate inflammation, with prostate total and transitional zone volume assessed using ultrasound (Wald = 17.6 and 12.5, both P < 0.0001), with arterial peak systolic velocity (Wald = 9.57, P = 0.002), with texture nonhomogeneity (hazard ratio (HR) = 1.87 (1.05-3.33), P < 0.05), with calcification size (Wald = 3.11, P< 0.05), but not with parameters of seminal vesicle size or function. In conclusion, in males of infertile couples, MetS is positively associated with prostate enlargement, biochemical (sIL8) and ultrasound-derived signs of prostate inflammation but not with prostate-related symptoms, which suggests that MetS is a trigger for a subclinical, early-onset form of benign prostatic hyperplasia.

INTRODUCTION

Metabolic syndrome (MetS) is a diagnostic category based on a cluster of abnormalities (abdominal obesity, impaired glucose metabolism, hypertriglyceridemia, reduced high-density lipoprotein (HDL) cholesterol and hypertension) that identifies subjects at high risk for diabetes and cardiovascular diseases.1234 Various diagnostic criteria for MetS have been proposed over recent years. Indeed, the parameters to be used for the diagnosis of MetS and their thresholds remains under debate.45 Nonetheless, insulin resistance has been recognized as the common feature underlying MetS,6 with visceral adiposity being the driving force.2345 In addition to diabetes and cardiovascular diseases, several other pathologic conditions are associated with MetS including nonalcoholic fatty liver disease, polycystic ovarian syndrome, obstructive sleep apnea and lipodystrophy.2 In addition, in the male, hypogonadism, erectile dysfunction and psychological disturbances are often associated with MetS.4789

A possible association between MetS and male infertility has also been hypothesized.1 Although several studies have evaluated the impact of being overweight and obesity on male reproductive health,10111213 only a few studies have assessed the influence of some MetS components on male fertility1 and none have considered MetS to be a diagnostic category. We evaluated this topic specifically in a cohort of 351 males of infertile couples without known genetic abnormalities.9 Even after adjusting for confounders, we noted that an increase in the number of MetS components associated with hypogonadism, poor sperm morphology (but not other sperm parameters), testis ultrasound nonhomogeneity, erectile dysfunction, somatization (expression of physical symptoms in the absence of medically explained physical illness) and depressive traits.9

In addition, in this last decade, a growing body of evidence has also documented an independent association between benign prostatic hyperplasia (BPH)/lower urinary tract symptoms (LUTS) and obesity/MetS.1415161718 In particular, we previously reported a positive correlation between MetS and BPH-related chronic inflammation in patients undergoing surgery for BPH.1920 A recently published epidemiological survey of the Boston area (BACH) confirmed an association between MetS and LUTS; however, when subjects were stratified by age, the association was confirmed only in the youngest individuals.21

This study is aimed at investigating systematically the possible associations between MetS and prostate-related symptoms and signs in a cohort of young men in infertile unions and to establish whether these associations correlate with fertility.

MATERIALS AND METHODS

We retrospectively evaluated a consecutive series of 187 male patients (age: 36.5 ± 8.3 years) who attended our Outpatient Clinic initially between January 2010 and December 2011 seeking medical care for infertility. Based on guidelines of the World Health Organization, infertility was defined as the inability of a sexually active couple to achieve pregnancy despite unprotected intercourse over a period of greater than 12 months.22 Subjects with karyotypic abnormalities (n = 3), chromosome Y microdeletions (n = 2) or an absence of at least one vas deferens and/or one seminal vesicle (n = 11) were excluded from the analysis. Hence, a cohort of 171 selected patients was used for the analyses. The sociodemographic and clinical phenotype of the sample population is summarized in .

All patients were evaluated prior to any treatment. All enrolled patients underwent the typical diagnostic protocol used for infertility in newly referred subjects at the Andrology Outpatient Clinic. They underwent a complete andrological and physical examination and blood pressure (mean of three measurements taken 5 min apart in the sitting position using a standard sphygmomanometer), height, weight and waist circumference were measured. In addition, routine scrotal and transrectal ultrasounds were performed because our regional healthcare system does not allow any genetic analysis on infertile patients unless a suspected obstruction has been evaluated. All data were collected as part of the routine clinical procedure; therefore, based on Italian law, approval from the local Ethical Committee was not required. In addition, at the time of the initial visit, all patients provided written, informed consent to have their clinical records included in a dedicated database to be used, anonymously, for clinical research purposes.

MetS assessment

MetS was defined, based on the National Cholesterol Education Program Third Adult Treatment Panel,26 as the presence of three or more of the following five factors: central obesity (waist circumference >102 cm), elevated triglycerides (≥1.7 mmol l-1 or treated for elevated triglycerides), elevated blood pressure (systolic blood pressure ≥130 mmHg and/or diastolic blood pressure ≥85 mmHg or treated for hypertension), elevated fasting glucose (≥6.1 mmol l-1 or treated for diabetes) and reduced HDL cholesterol (<1.03 mmol l-1 or treated for dyslipidemia).

Biochemical parameters

Blood samples were drawn in the morning after an overnight fast to determine blood glucose (using the glucose oxidase method; Aeroset Abbott, Rome, Italy), HDL cholesterol and triglycerides (using the automated enzymatic colorimetric method; Aeroset Abbott, Rome, Italy), total testosterone (TT, using the electrochemiluminescent method; Modular Roche, Milan, Italy), insulin levels and sex hormone binding globulin (SHBG) using an electrochemiluminescence immunoassay (Roche Diagnostics, Mannheim, Germany). Free testosterone was calculated based on Vermeulen's formula (available at http://www.issam.ch/freetesto.htm).27

Semen analyses and determination of seminal plasma interleukin 8 (sIL-8) levels

On the same day as the ultrasound, all patients underwent semen analysis, which was performed based on World Health Organization criteria.28 In addition, routine urine and seminal cultures were assessed in all men. Furthermore, sIL-8, a reliable surrogate marker of prostatitis,29 was also quantified. Seminal plasma aliquots were frozen and stored for later quantification of sIL-8 levels using conventional two-site enzyme-linked immunosorbent assay (ELISA; human IL-8 ELISA set; BD Biosciences, San Diego, CA, USA) according to the manufacturer's instructions.29 Each seminal plasma sample was diluted from 1:5 to 1:625. Assay sensitivity for sIL-8 was <1 pg ml-1.

Screening of prostate-related symptoms and lower urinary tract symptoms

Patients were asked to complete the Italian translation of the National Institutes of Health Chronic Prostatitis Symptom Index (NIH-CPSI),30 which is a brief self-reported questionnaire for screening prostatitis symptoms and that scores for pain, voiding symptoms and quality of life. The NIH-CPSI total score was calculated as the sum of the scores of these domains. LUTS were evaluated using the Italian translation of the International Prostate Symptom Score (IPSS), which is a brief self-administered questionnaire for screening symptoms related to BPH and that comprises seven questions on symptoms and one question on quality of life.31

Transrectal color Doppler ultrasonography (CDU)

We previously reported an association between MetS and scrotal parameters in a larger cohort of subjects who attended our unit for infertility.9 The characteristics of that cohort were not different from the cohort in this study; thus, in this study, we focused on the possible associations between MetS and transrectal ultrasound features.

All patients underwent scrotal and transrectal CDU; the latter before and after ejaculation. To prevent bias on the part of the examiner, scrotal and transrectal CDU was performed intermittently by two experienced physicians who were unaware of the clinical data and who used the same ultrasonographic console (Hitachi H21, Hitachi Medical System, Tokyo, Japan). Prostate and seminal vesicle CDU features were studied by scanning the organs at 5 mm intervals at various longitudinal, transverse and oblique scans based on previous studies2324253233 using a transrectal biplanar probe (linear transducer U533L 7.5 MHz; convex transducer U533C 6.5 MHz) and an ‘end fire’ probe (V53W 6.5 MHz, field of view 50°–200°). Based on: 1) anteroposterior (APD) and 2) transverse (TD) diameters at the maximal dimensions and 3) the superior-inferior or longitudinal diameter (LD) at the maximal length from the base to the apex of the midline sagittal plane, prostate volume was calculated using the formula of the ellipsoid (APD × TD × LD × π/6) based on previous publications.233435 During this exam, similar methods were also used to determine the transitional zone volume (TZV) based on a previous publication.35 Prostate echogenicity was defined based on previous studies.23243233 The severity of prostate texture nonhomogeneity was classified using an arbitrary Likert scale and scores of 0–3 (0 = homogeneous texture, 1 = mild, 2 = moderate and 3 = severe texture nonhomogeneity). Prostate vascularization, prostate hyperemia and arterial prostatic peak systolic velocity (APPSV) were defined based on previous studies23243637 and evaluated prior to ejaculation to avoid post-ejaculatory changes in vascular flow pattern as has been previously reported.38 Seminal vesicle ultrasound features and abnormalities were defined based on previous studies.2325333940 Ejaculatory duct CDU characteristics were evaluated after ejaculation to better emphasize indirect CDU signs of subobstruction.232539

Scrotal CDU was performed systematically using a 7.5 MHz high-frequency linear probe (L54M 6–13 MHz) at various longitudinal, transverse and oblique scans with patients lying in a supine position.233233 Testicular and epididymal CDU features were examined based on previous studies.233233

Identification of case patients and controls

MetS was defined as described above. Subjects with ≥3 MetS components (n = 22) were compared with controls selected from the same cohort at a 1:2 ratio (n = 44). For each case, the first two patients following those with MetS within the same series who were the same age (±4 years) and who showed a similar TT level (± 4 nmol l−1), smoking habit (current/nonsmoker) and moderate-severe alcohol consumption (current/no consumption of ≥4 drinks per day based on a previous publication41). For statistical analyses that compared cases with age-, TT-, smoking habit-, moderate-severe alcohol consumption-matched controls, associations with P < 0.05 were considered significant.

Data analyses

Data were expressed as the mean ± s.d. when normally distributed, the median (quartiles) for parameters with non-normal distributions and as percentages when categorical. Correlations were assessed using Spearman's or Pearson's methods as appropriate. Differences between more than two groups were assessed using one-way analysis of variances. Unpaired two-sided Student's t-test was used to compare means of normally distributed parameters. Relative risks and 95% confidence intervals were calculated for correlations of categorical parameters, and chi-squared tests were used for comparisons. Stepwise multiple linear, logistic binary or ordinal regressions were applied for multivariate analyses as appropriate. All statistical analyses were performed in SPSS (Statistical Package for Social Sciences, Chicago, USA) for Windows 20.0.

RESULTS

Among the 171 patients studied (age: 36.6 ± 8.4 years), 44.4% (n = 76) showed no components of MetS, whereas one, two, three, four and five MetS factors were present in 47 (27.5%), 26 (15.2%), 16 (9.4%), three (1.75%) and three (1.75%) subjects, respectively. Twenty-two subjects (12.9%) fulfilled the criteria of National Cholesterol Education Program Third Adult Treatment Panel MetS. Subjects with MetS were older (43.8 ± 10.6 vs 35.5 ± 7.5 years for MetS and no-MetS subjects, respectively; P < 0.0001). No difference in the percentage of subjects who smoked currently or consumed moderate-severe amounts of alcohol was found when MetS and no-MetS subjects were compared (21.1% vs 28.6%, P = 0.496; 16.7% vs 24.8%, P = 0.233; MetS vs no-MetS subjects, respectively). In addition, no difference in the prevalence of leukocytospermia or current positive urine and/or seminal cultures was observed when men with and without MetS were compared (5.3% vs 8.8%, P = 0.508; 5.6% vs 8.1%, P = 0.575; MetS vs non-MetS subjects, respectively).

Correlations of MetS with hormonal, clinical and semen parameters

Age-adjusted logistic ordinal models showed that insulin levels increased as a function of MetS components (Wald = 29.5 (0.09–0.18), P < 0.0001) (). Figure also showed the age-adjusted relationships between insulin and TT, SHBG and calculated free testosterone: all of these parameters decreased as a function of increasing insulin levels (adjusted r = -0.359, P < 0.0001; adjusted r = -0.200, P < 0.001; adjusted r = -0.320, P < 0.0001, respectively). In view of these associations, all the following analyses were adjusted for age, insulin and TT levels.

Association between insulin levels and the number of metabolic syndrome (MetS) components (National Cholesterol Education Program Third Adult Treatment Panel classification) (a), total testosterone (b), sex hormone binding globulin (c) and calculated free testosterone levels (d). In (a), subjects with no, one or more MetS components are indicated. In (b–d), subjects with or without MetS are shown as filled or empty dots, respectively.

Digitorectal examinations revealed that an enlarged prostate was positively associated with the number of MetS components (hazard ratio (HR) =1.47 (1.03–2.14)) for each increment in the number of MetS components, P < 0.05). Of the semen parameters, only normal semen morphology was negatively associated with an increasing number of MetS components (Wald = 5.59 (-0.15 to -0.01), P < 0.02).

No association between MetS components and other physical or seminal parameters including leukocytospermia was observed (data not shown). Finally, no association between MetS components and current positive urine and/or seminal cultures was detected (data not shown).

Correlations of MetS with prostate-related symptoms and signs (sIL-8)

No association was found between MetS and prostate-related symptoms as captured by both NIH-CPSI and IPSS (not shown). A stepwise, positive correlation between the number of MetS components and sIL-8 levels was observed (Wald = 4.32 (0.04–1.48), P < 0.05) (). In particular, among the MetS components, only waist circumference was positively associated with sIL-8 ().

Association between metabolic syndrome (MetS) and seminal interleukin 8 (sIL-8) levels or prostate volume at ultrasound. Association between the number of MetS components (NCEP-ATPIII classification) and sIL-8 levels (a) or prostate volume at ultrasound (c). The number of subjects with no, one or more MetS components is indicated. Hazard ratio (95% confidence interval) for sIL-8 levels (b) and prostate volume at ultrasound (d) as detected by iterative logistic regression analyses considering MetS components as putative predictors. MetS components are defined by abnormal parameters or by specific therapy (t) based on NCEP-ATPIII groupings. HDL, high density lipoprotein.

Correlations of MetS with transrectal ultrasound parameters

A progressively higher prostate volume was detected by ultrasound as a function of an increasing number of MetS components (Wald = 17.6 (0.05–0.13), P < 0.0001) (). In a logistic iterative analysis, of the MetS factors, waist size and reduced HDL cholesterol were significantly associated with prostate volume (). Interestingly, TZV was positively correlated with total prostate volume even after adjusting for the aforementioned confounders (adjusted r = 0.757, P < 0.0001). Thus, TZV also increased as a function of an increasing number of MetS components (Wald = 12.5 (0.07–0.24), P < 0.0001). In addition, TZV of MetS subjects was significantly higher than other subjects (10.7 ± 10.8 vs 3.9 ± 2.6 ml; P < 0.0001; subjects with MetS vs those without MetS, respectively). Finally, similarly to total prostate volume results, TZV was significantly associated with reduced HDL cholesterol levels (HR = 1.15 (1.01–1.31), P < 0.05). Conversely, only a trend towards statistical significance was observed for increased waist (HR = 1.11 (0.99–1.24), P = 0.08).

A positive association between an increasing number of MetS components was also observed for some ultrasonographic prostate features, such as arterial peak systolic velocity (APPSV, Wald = 9.57 (0.05–0.24), P = 0.002) () and diameter of the major calcification (Wald = 3.11 (0.01–0.13), P < 0.05) (). Using a binary logistic model, moderate-severe prostate texture nonhomogeneity was also associated with an increase in the number of MetS components (HR = 1.87 (1.05–3.33) for each increment in the number of MetS components, P < 0.05) (). Of the MetS components, only increased waist circumference was significantly associated with APPSV () and with moderate-severe texture nonhomogeneity (). After adjusting for confounders, no associations between MetS related-prostate CDU abnormalities and standard semen parameters were observed (). Finally, no associations between the number of MetS components and seminal vesicle features or the mean diameter of the deferential ampulla were observed (data not shown).

Association between metabolic syndrome (MetS) and arterial prostatic peak systolic velocity (APPSV), texture nonhomogeneity and major calcification size of the prostate as evaluated using color Doppler ultrasound (CDU). Association between the number of metabolic syndrome (MetS) components (NCEP-ATPIII classification) and APPSV (a), moderate-severe nonhomogeneity prevalence (c) or major calcification size (c, inset) of the prostate as evaluated using CDU. The number of subjects with no, one or more MetS components is indicated. Hazard ratio (95% confidence interval) for APPSV (b) and prostate moderate-severe nonhomogeneity (d) as detected by iterative logistic regression analysis considering MetS components as putative predictors. MetS components are defined by abnormal parameters or by specific therapy (t) based on NCEP-ATPIII groupings. m-s, moderate-severe.

Case-control analyses

Correlations between MetS and seminal or ultrasound parameters showing statistical significance were further assessed by comparing subjects with MetS with matched controls (matched for age, TT, smoking habit and moderate-severe alcohol consumption) () at a 1:2 ratio. Even in the case-controlled analyses, subjects with three or more MetS components showed a lower percentage of normal sperm morphology, higher sIL8 levels and more frequent prostate abnormalities, such as greater volume, higher arterial peak systolic velocity, greater calcification size and prevalence of moderate-severe nonhomogeneity (). No difference in the prevalence of leukocytospermia or current positive urine and/or seminal cultures was observed when MetS subjects were compared with controls ().

Correlations of insulin levels with clinical, seminal and transrectal ultrasound parameters

Univariate analyses revealed positive associations between insulin levels and increased prostate volume detected by either digitorectal examination (RR = 1.08 (1.03-1.14), P = 0.002 for each insulin mU l-1 increment) or ultrasound (r = 0.294, P < 0.0001), moderate-severe texture nonhomogeneity (RR = 1.08 (1.00-1.17), P = 0.05 for each insulin mU l-1 increment), hyperemia (RR = 1.05 (1.00-1.11), P < 0.05 for each insulin mU l-1 increment) and arterial peak systolic velocity (r = 0.286, P < 0.0001) and a negative correlation with the prostatic venous plexus diameter (r = -0.180, P < 0.02). When a multivariate regression model was applied that included age and TT, the significant associations between insulin levels and prostate volume detected both by digitorectal examination (HR = 1.07 (1.01-1.13), P < 0.05 for each insulin mUl-1 increment) and ultrasound (adjusted r = 0.327, P < 0.0001) () and arterial peak systolic velocity (adjusted r = 0.229, P = 0.002) () were confirmed. However, when the number of MetS components was introduced into the same model, only prostate volume detected by ultrasound was significantly associated with insulin levels (adjusted r = 0.171, P < 0.05).

Associations between insulin levels and prostate volume (a) or arterial prostatic peak systolic velocity (b) evaluated using color Doppler ultrasound (CDU). Subjects with or without metabolic syndrome are shown as filled or empty dots, respectively

Even after adjusting for the aforementioned confounders, no associations between insulin levels and semen parameters, sIL-8 levels, seminal vesicle features or diameters of the deferential ampulla were observed (data not shown).

DISCUSSION

This study demonstrates that in a cohort of relatively young male subjects examined for infertility, a stepwise, component-dependent association was observed between an increase in MetS severity and prostate enlargement and/or inflammatory signs (including sIL-8 levels and CDU abnormalities), but not with current infection of the male genital tract. No association between MetS-related prostate CDU abnormalities and semen parameters was detected. However, in this cohort, MetS was associated with poor sperm morphology. Reduced HDL levels and increased abdominal adiposity were the main correlates of prostate enlargement in this young, asymptomatic cohort.

The association between MetS and prostate enlargement is consistent with several previous reports.204243444546474849 Thus far, only a few studies have examined relatively young adults, and they offer conflicting results.5051 Here, we report a novel association in a relatively young population (mean age 36.6 ± 8.4 years) of males of infertile couples. Increased central obesity and reduced HDL cholesterol were the parameters that most closely correlated with prostate enlargement. A potential relationship between BPH/prostate enlargement and obesity or increased waist circumference has been widely reported in several,1452 but not all previous studies.20 Moreover, low HDL cholesterol has been previously reported as a risk factor for the development of BPH.2042434445 Our data suggest that MetS, and particularly high waist circumference and reduced HDL cholesterol, may play an important role in prostate growth onset at a young age.

We also noted a significant, stepwise correlation between the number of MetS components and seminal IL-8 (sIL-8), which has been proposed as a surrogate marker of prostate inflammation.29535455 IL-8 is a proinflammatory chemokine that is secreted by several cell types and that contributes to inflammation by acting in concert with IL-1β and IL-6.565758 Higher IL-8 levels have been reported in the expressed prostatic secretions of subjects with BPH, bacterial prostatitis and chronic prostatitis/chronic pelvic pain syndrome (CP/CPPS).596061 Because IL-8 in seminal plasma is considerably higher when compared with serum levels,6263 local (within the male genital tract) production has been suggested.555963 Of different cytokines and chemokines, sIL-8 appears to be the most reliable and predictive surrogate marker of prostatitis,2954 and it is associated with CDU features suggestive of prostate inflammation in patients with male accessory gland infection (MAGI).23 IL-8 is actively involved in BPH-associated chronic inflammation and mediates epithelial and stromal cell proliferation.54 In BPH tissues, epithelial and stromal cells secrete IL-8 actively2964 in response to varying stimuli including the proinflammatory cytokines interferon (IFN) γ and IL-17 that are produced by prostate-infiltrating Th1 and Th17 cells, respectively.53656667 In particular, human stromal prostatic cells actively contribute to the organ-specific inflammatory process by acting as targets of bacterial or viral toll-like receptors agonists and as antigen-presenting cells capable of activating antigen-specific CD4 + T cells.68 In BPH cells, toll-like receptor activation leads to the production of proinflammatory cytokines (IL-6) and chemokines (IL-8 and CXCL10) capable of recruiting CXCR1 and CXCR2-positive leukocytes and CD15 + neutrophils.29 Finally, IL-8 stimulates overgrowth of prostate stromal and epithelial cells by directly promoting the proliferation of senescent epithelial cells,69 stromal transdifferentiation of myofibroblasts70 and by increasing secretion of fibroblast growth factor 2.71 Hence, IL-8 appears to be the link between T cell-mediated inflammatory responses and cell proliferation in the pathogenesis of BPH.5467

A higher prevalence of MetS components was also associated with other CDU features of prostate inflammation including texture nonhomogeneity, major calcification size and elevated APPSV. Increased waist size is the common determinant of all of these CDU abnormalities. Prostate nonhomogeneity is typically considered a CDU abnormality related to inflammation32 and has been previously associated with elevated sIL-8 levels in infertile subjects with MAGI.23 Moreover, prostatic hyperechogenicity, which is associated with areas of calcification, has been previously proposed to be a CDU feature suggestive of MAGI.33 In particular, prostatic high-density echoes are considered the sonographic correlates of prostatic calculi and corpora amylacea, which as confirmed by histology performed on ultrasound-guided biopsies of the prostate.72 A recent report indicated that prostatic calculi and corpora amylacea comprised acute inflammatory proteins including lactoferrin, calprotectin, myeloperoxidase and α-defensins, all of which are in neutrophil granules.73 Prostatic calcifications are common in patients with CP and have been associated with the maintenance or enhancement of prostate inflammation, bacterial colonization and duration of symptoms.7475 A positive correlation between sIL-8 levels and calcification size has also been previously reported by our group.23 Finally, we noted that APPSV correlated with MetS severity. Arterial PSV reflects tissue inflammation at various sites including the thyroid,7677 exocrine glands7879 and synovial membrane/joints.8081 Thus far, APPSV has been studied for varying purposes including evaluation of BPH,82 prostate cancer,83 varicocele-related prostate CDU changes84 and premature ejaculation.37 More recently, elevated APPSV has been proposed as a CDU parameter that correlates with prostate inflammation,232437 and in the prostate, APPSV is closely related to sIL-8.23

No correlation was found between the number of MetS components and current positive urine and/or seminal cultures suggesting that MetS is not associated with current infection of the male genital tract but is rather associated with chronic inflammation.

The relationship between central obesity and dyslipidemia with prostate overgrowth and inflammation, even in young subjects, is the main finding of this study. We previously reported a clear-cut association between MetS severity and prostate size20 and inflammation1985 in cohorts of aged subjects. In the study of Gacci et al.,20 reduced HDL cholesterol and increased triglyceride levels were also noted to be the main determinants of MetS-related prostate alterations. However, previous studies were performed in old individuals undergoing surgery for BPH.1985 By culturing BPH stromal cells obtained from those patients, we demonstrated that in addition to tumor necrosis factor-α and lipopolysaccharide, oxidized LDL was capable of increasing IL-8, IL-6 and basic fibroblast growth factor secretion.19 In addition, tumor necrosis factor-α sensitized BPH cells to oxidized LDL by inducing its receptor (LOX-1)19. In a rabbit model of MetS we recently demonstrated that a 3-month high-cholesterol diet (HFD) induced severe prostatic inflammation characterized by increased corpora amylacea, fibrosis and hypoxia.86 In addition, the mRNA expression of several proinflammatory cytokines including IL-8 and T lymphocyte, macrophage, neutrophil and fibrosis/myofibroblast activation markers were significantly increased in the prostate of HFD animals.86 Together, all of these data suggest that BPH may be viewed as a complex disorder that also involves a metabolic component that may begin early in the life of the male, and although asymptomatic, it is likely detectable even in the early stages of the disease as suggested by this study. The mechanisms underpinning the relationship between MetS and prostate inflammation are likely to be similar in young and old men, but chronic exposure to elevated inflammation may contribute to BPH in the long-term.

Because hyperinsulinemia and insulin resistance represent the cornerstone of all definitions of MetS,36 all of the data reported here were adjusted for insulin levels. When the specific contribution of hyperinsulinemia was considered, after adjusting for MetS components and TT levels and age, we observed a specific effect of increased insulin levels only on prostate volume and not on prostate inflammation. This finding is in apparent contrast with results published recently by our group that showed that insulin increased IL-8 release from myofibroblastic hBPH cells.19 However, in that study, the effect of insulin was negligible when compared with oxidized LDL (sixfold lower).19 The growth-promoting activity of insulin on the prostate gland is well-documented in several experimental and epidemiological studies.87

In this study, no association between an increase in the number of MetS components and prostate-related symptoms was observed, using either NIH-CPSI or IPSS scores. The lack of correlation between MetS and LUTS is consistent with some studies, but contrasts with most previous studies.151618 However, all previous studies were performed in aged cohorts; whereas, our data were obtained from young subjects with relatively small prostates as assessed by ultrasound.

Finally, even after adjusting for confounders including the hypogonadal status, we observed an association between the number of MetS components and poor sperm morphology. These results confirmed previous findings obtained using a larger cohort of males of infertile couples where we assessed the possible correlations of MetS with scrotal parameters.9 The possible impact of MetS on sperm morphology was discussed in detail in that study. Here, we extend our investigation to possible associations between MetS-related prostate CDU features and semen parameters and note no correlation. This finding suggests that the effect of MetS on sperm quality is independent of MetS-related prostate abnormalities. Although a possible association between CP or MAGI and sperm quality has been proposed and some prostate CDU features have been proposed as suggestive of MAGI,33888990 a specific association between prostate CDU features and sperm parameters alterations has not been demonstrated. Here, we report that a number of MetS components, but not related prostate CDU abnormalities, are associated with poor sperm morphology.

This study has several limitations. First, the results were derived from patients consulting an Italian Andrology Clinic for infertility, and our population could have different characteristics from the general male population or those males consulting general practitioners for reasons other than infertility. Second, a true control group comprising age-matched, apparently healthy, fertile men is lacking and therefore true normative data on sonographic parameters cannot be inferred. Third, the data cannot distinguish between men who are clinically infertile and those who are not, and whether this feature confounds the results is unknown. Fourth, this study relies on a relatively small and young sample. Fifth, markers of prostate inflammation other than sIL-8 are lacking. Furthermore, parallel mechanistic studies to determine the consequences of IL-8 secretion on prostate growth and differentiation are warranted. Finally, the link between MetS or insulin levels and prostate enlargement in this study are correlative only. Statistically significant associations in a cross-sectional study do not infer causality.

However, this study also has several strengths. First, this study systematically evaluates several hormonal, seminal, laboratory and ultrasound parameters in a consecutive series of males of infertile couples. Second, during the same sonographic session, this study examined both scrotal and transrectal ultrasound features before and after ejaculation. Third, this study was performed on a sample of relatively young, infertile men and investigated a population that is poorly studied in the scientific literature. Fourth, this study considers several possible confounders, such as age, testosterone and insulin levels, smoking habits, alcohol consumption, leukocytospermia and positive semen and urine cultures of the patients. Fifth, a statistical analysis comparing patients with MetS with age-, TT-, smoking habits- and alcohol consumption-matched controls was performed. Finally, the study examined several end points simultaneously within the same population, which enabled a valid comparison of the coprevalence of the examined parameters and supported their possible association with the number of MetS components.

CONCLUSIONS

This study demonstrates that in a cohort of men with infertility, a component-dependent, stepwise association was observed between an increase in the number of MetS components and the total and transitional zone prostate enlargement and prostate related-inflammatory signs, but not symptoms or current infection of the male genital tract, which suggests a subclinical inflammation of the prostate. Relative prostate overgrowth may also correlate with MetS-related hyperinsulinemic state. In addition, MetS but not MetS's related prostate CDU abnormalities was associated with poor sperm morphology.

AUTHOR CONTRIBUTIONS

FL, GC and MM conceived and designed the study, performed the statistical analysis and drafted the manuscript. LV, MG and GF participated in the design of the study. FL, EM, MR and SC participated in data collection and management. All authors read and approved the final manuscript.

COMPETING INTERESTS

The authors declare no competing interests.

Table 1

Clinical and transrectal color Doppler ultrasound characteristics of the sample

Table 2

Univariate associations between MetS-related prostate color Doppler ultrasound characteristics and semen parameters

Table 3

Comparisons between subjects with metabolic syndrome and 1:2 ratio-matched controls (matched for age, total testosterone, smoking habits and moderate-severe alcohol consumption)