Metabolic syndrome (MetS) is a diagnostic category which identifies subjects at high risk for diabetes and cardiovascular diseases, erectile dysfunction (ED) and male hypogonadism. However, MetS impact on male infertility has been poorly studied. We systematically evaluated possible associations between MetS and clinical characteristics in men with couple infertility. Out of 367 consecutive subjects, 351 men without genetic abnormalities were studied. MetS was defined according to the International Diabetes Federation&American Heart Association/National Heart,Lung, and Blood Institute classification. All men underwent physical, hormonal, seminal and scrotal ultrasound evaluation. Erectile and ejaculatory functions were assessed by International Index of Erectile Function-15 erectile function domain (IIEF-15-EFD) and Premature Ejaculation Diagnostic Tool (PEDT), respectively, while psychological symptoms by Middlesex Hospital Questionnaire. Out of 351 patients, 27 (7.7%) fulfilled MetS criteria. Among ultrasound features, in an age-adjusted logistic model, only testis inhomogeneity was significantly associated with increasing MetS factors (HR = 1.36 [1.09–1.70]; p < 0.01). In an age-adjusted model, MetS was associated with a stepwise decline in total testosterone (TT) (B = −1.25 ± 0.33; p < 0.0001), without a concomitant rise in gonadotropins. At univariate analysis, progressive motility and normal morphology were negatively related to the number of MetS components (both p < 0.0001), but when age and TT were introduced in a multivariate model, only sperm morphology retained a significant association (B = −1.418 ± 0.42; p = 0.001). The risk of ED (IIEF-15-EFD score <26) increased as a function of the number of MetS factors, even after adjusting for age and TT (HR = 1.45[1.08–1.95]; p < 0.02). No association between PEDT score and MetS was observed. Finally, after adjusting for age and TT, somatization and depressive symptoms were associated with increasing MetS components (B = 0.66 ± 0.03, p < 0.05; B = 0.69 ± 0.03, p < 0.02; respectively). In conclusion, in men with couple infertility, MetS is associated with hypogonadism, poor sperm morphology, testis ultrasound inhomogeneity, ED, somatization and depression. Recognizing MetS could help patients to improve not only fertility but also sexual and overall health.
Metabolic syndrome (MetS) represents a constellation of abnormalities, including abdominal obesity, impaired glucose metabolism, dyslipidaemia (hypertriglyceridaemia, low HDL cholesterol) and hypertension, which identifies subjects at high risk for diabetes and cardiovascular diseases (Cornier et al., 2008; Kasturi et al., 2008; Eckel et al., 2010; Corona et al., 2011a).
Several other pathological conditions are associated with MetS, including non-alcoholic fatty liver disease, polycystic ovarian syndrome, obstructive sleep apnoea, lipodystrophy and microvascular disease (Cornier et al., 2008). In addition, in men, hypogonadism, erectile dysfunction and psychological disturbances are also often comorbid with MetS (Corona et al., 2006, 2011a,b). Although a possible association between MetS and male infertility has been hypothesized (Kasturi et al., 2008), it has never been demonstrated.
Different diagnostic criteria for MetS have been proposed over the years. In fact, the definition of parameters to be used for the diagnosis of MetS, as well as their thresholds, is under constant debate (Corona et al., 2009a, 2011a). Nonetheless, insulin resistance and increased adiposity have been universally recognized as the main features underlying MetS (Cornier et al., 2008; Corona et al., 2009a, 2011a), although there is increasing evidence for other key components of the syndrome, such as oxidative stress, proinflammatory state or exaggerated sympathetic nervous system activation (Cornier et al., 2008). The impact of overweight or obesity on the reproductive health of men has been only partially elucidated. In fact, although scientific interest in this issue has grown in the last years (see for review Hammoud et al., 2008; MacDonald et al., 2010; Sermondade et al., 2012), available clinical studies investigating the relationship between obesity and male infertility are relatively poor compared to the extensive body of research investigating its association with female infertility (Pasquali et al., 2007; Zain & Norman, 2008; Loret de Mola, 2009; Wilkes & Murdoch, 2009; Brewer & Balen, 2010; Rittenberg et al., 2011). Hence, while the relationship between obesity and female infertility is more than clear, the association with male infertility is still up for discussion (Hammoud et al., 2008; MacDonald et al., 2010). In addition, a recent meta-analysis revealed little evidence for a relationship between obesity and poor semen parameters, despite a clear association with hypogonadism (MacDonald et al., 2010). However, a more recent meta-analysis reported that overweight and obesity are both associated with an increased risk of azoospermia or oligozoospermia (Sermondade et al., 2012).
Available animal models of MetS, both genetically (Vartanian et al., 2006) and high fat diet- (Mallidis et al., 2011) induced, suggest that MetS is associated with deranged spermatogenesis and poor sperm quality.
As MetS is essentially based on increased adiposity and it is associated with male hypogonadism (Corona et al., 2011a), erectile dysfunction and psychological disturbances (Corona et al., 2006), and all these factors might, in different ways, affect reproductive capacity (Hämmerli et al., 2009), we investigated their possible correlations with MetS. This study is, in fact, aimed at evaluating possible associations between MetS, semen and hormonal parameters, as well as clinical characteristics, including sexual, ultrasound and psychological characteristics, in a cohort of men with couple infertility.
Materials and methods
We studied a consecutive series of 376 male patients (mean age 36.0 ± 7.5 years) attending our Outpatient Clinic for the first time from January 2008 to December 2011, seeking medical care for couple infertility. Couple infertility was defined as the inability of a sexually active couple to achieve pregnancy despite unprotected intercourse for a period greater than 12 months, according to the World Health Organization (WHO, 2000). Subjects with karyotype abnormalities (n = 3), chromosome Y microdeletions (n = 3) and uni- or bilateral (n = 4 and n = 15 respectively) absence of vas deferens were excluded from the analysis. The socio-demographic and clinical characteristics of the sample are summarized in Table 1.
Table 1. Clinical and scrotal colour-Doppler ultrasound characteristics of the whole sample and of subjects with or without metabolic syndrome
All patients n = 351
Without MetS n = 324
With MetS n = 27
Data are expressed as mean ± standard deviation or as median (quartiles), when appropriate, and as percentage, when categorical.
HDL: high density lipoprotein, BP: blood pressure, FSH: follicle stimulating hormone, LH: luteinizing hormone, PRL: prolactin, TSH: thyroid stimulating hormone, Tp: therapy. Echographic defined severe varicocoele = basal venous reflux increasing after Valsalva's manoeuvre at sonography (according to Isidori & Lenzi, 2008; Lotti et al., 2009).
Clinical and laboratory parameters
36.0 ± 8.0
35.5 ± 7.4
43.7 ± 10.4
Current smoker (%)
Current alcohol consumption (≥2 drinks/die) (%)
Diabetes mellitus (%)
95.5 ± 11.6
94.1 ± 9.9
111.4 ± 16.7
4.97 ± 0.68
4.89 ± 0.64
5.56 ± 0.75
HDL cholesterol (mmol/L)
1.28 ± 0.33
1.32 ± 0.32
1.06 ± 0.35
Systolic BP (mm Hg)
124.7 ± 12.3
123.8 ± 11.2
135.0 ± 18.5
Diastolic BP (mm Hg)
80.0 ± 7.6
79.5 ± 7.1
85.2 ± 10.2
Mean testis volume (Prader) (mL)
18.4 ± 4.9
18.3 ± 4.8
19.7 ± 5.4
History of cryptorchidism (%)
Total testosterone (nmol/L)
16.4 ± 6.3
16.7 ± 6.2
13.8 ± 6.5
Sex hormone binding globulin (nmol/L)
Calculated free testosterone (nmol/L)
0.344 ± 0.120
0.349 ± 0.117
0.299 ± 0.139
Central obesity (waist circumference > 102 cm)
Elevated fasting glucose (≥5.6 mmol/L) or tp
Elevated triglycerides (≥1.7 mmol/L) or tp
Reduced HDL cholesterol (<1.03 mmol/L) or tp
Elevated BP (BP ≥ 130/85 mm Hg) or tp
Azoospermic subjects, %
Normozoospermic subjects, %
Sexual abstinence (days)
4.2 ± 1.9
4.2 ± 2.0
4.0 ± 1.3
7.5 ± 0.3
7.5 ± 0.3
7.6 ± 0.2
Semen volume (mL)
Sperm concentration, × 106/mL
Spermatozoa per ejaculate, × 106/mL
Sperm progressive motility (%)
36.5 ± 20.4
36.2 ± 20.7
39.3 ± 16.9
Sperm morphology, % normal forms
History of infertility
Duration of infertility (years)
1.9 ± 1.6
1.8 ± 1.5
2.3 ± 2.4
Female partner age (years)
33.8 ± 6.4
33.7 ± 6.4
35.9 ± 6.7
Colour-Doppler ultrasound parametersTestis
Mean testis volume (mL)
14.6 ± 4.8
14.5 ± 4.6
15.8 ± 5.9
Epididymis and vas deferens
Mean size of the head (mm)
9.1 ± 2.1
9.0 ± 1.8
10.0 ± 4.1
Mean size of the tail (mm)
4.4 ± 1.3
4.3 ± 1.3
4.7 ± 1.3
Mean size of the vas deferens (mm)
3.9 ± 0.9
3.9 ± 0.9
4.1 ± 1.1
Coarse tail calcifications
All patients were evaluated before beginning any treatment. All patients enrolled underwent the usual diagnostic protocol applied to newly referred subjects at the Andrology Outpatient Clinic for infertility. All patients underwent a complete andrological and physical examination, with measurement of blood pressure (mean of three measurements 5 min apart, in sitting position, with a standard sphygmomanometer), height, weight and waist. In addition, scrotal ultrasound was routinely performed. All the data provided were collected as part of a routine clinical procedure and therefore, according to Italian law, approval from the local Ethical Committee was not required. In addition, at the time of the first visit, all patients gave their written informed consent to have their clinical records included in a dedicated database and they were aware that their data, after having been made anonymous, would be used for clinical research purposes.
Metabolic syndrome assessment
Metabolic syndrome has been defined according to the International Diabetes Federation and the American Heart Association/National Heart, Lung, and Blood Institute (IDF & AHA/NHLBI) (Alberti et al., 2009) as the presence of three or more of the following five factors: central obesity (waist circumference >102 cm), elevated triglycerides (≥1.7 mmol/L or on drug treatment for elevated triglycerides), elevated blood pressure (BP ≥ 130 mm Hg systolic blood pressure or 85 mm Hg diastolic blood pressure or on antihypertensive drug treatment in a patient with a history of hypertension), elevated fasting glucose (≥5.6 mmol/L or on drug treatment for elevated glucose) and reduced high density lipoprotein (HDL) cholesterol (<1.03 mmol/L or on drug treatment for reduced HDL cholesterol).
Evaluation of erectile function
Patients were asked to complete the EFD of the International Index of Sexual Function-15 (IIEF-15) (Cappelleri et al., 1999), in its Italian translation. An IIEF-15-EFD score of <26 indicates erectile dysfunction (ED) (Cappelleri et al., 1999).
Evaluation of premature ejaculation status
Patients were asked to complete the Premature Ejaculation Diagnostic Tool (PEDT) (Symonds et al., 2007), in its Italian translation. PEDT is a brief, multidimensional, psychometrically validated, 5-item, self-reported questionnaire for diagnosing ejaculatory status, which provides scores for five subdomains, #1 control, #2 frequency, #3 minimal stimulation, #4 distress and #5 interpersonal difficulty. PEDT score was calculated as the sum of the scores of these domains. A PEDT score of ≤8 indicates no PE (Symonds et al., 2007).
Screening of psychological traits
Patients were asked to complete the Middlesex Hospital Questionnaire, modified (MHQ) (Crown & Crisp, 1966), a brief self-reported questionnaire for the screening of mental disorders, which provides scores for free-floating anxiety (MHQ-A), phobic anxiety (MHQ-P), somatization (MHQ-S), obsessive-compulsive (MHQ-O), depressive (MHQ-D) and hysterical (MHQ-H) traits and symptoms.
All patients underwent scrotal Colour-Doppler ultrasonography (CDU), using the ultrasonographic console Hitachi H21 (Hitachi Medical System, Tokyo, Japan). To prevent bias on the part of the examiner, scrotal CDU was performed intermittently by two experienced physicians (F. L. and G. C.), unaware of the clinical data. The CDU characteristics of the sample are summarized in Table 1.
Scrotal CDU was performed in various longitudinal, transverse and oblique scans (Behre et al., 1995; Vicari, 1999; Lotti et al., 2011, 2012) using a 7.5 MHz high-frequency linear probe (L54M 6–13 MHz, Hitachi Medical Corporation, Tokyo, Japan). Testicular and epididymal CDU features were examined according to previous studies (Behre et al., 1995; Vicari, 1999; Isidori & Lenzi, 2008; Lotti et al., 2011, 2012; Migaleddu et al., 2012). In particular, testis inhomogeneity (striated pattern) was defined as previously reported (Cohn et al., 1996; Isidori & Lenzi, 2008; Loberant et al., 2010; Migaleddu et al., 2012) [see Fig. 1(A), inset].
All patients underwent colour-Doppler assessment of penile arteries in the flaccid state with a 7.5-MHz high-frequency linear probe (L54M 6–13 MHz), a pulsed Doppler investigation frequency of 5 MHz and a colour flow mapping capability, according to previous studies (Mancini et al., 2000; Corona et al., 2008a). The transducer was placed longitudinally on the ventral surface at the base of the penis. Colour flow mapping was helpful in obtaining an accurate angle corrected velocity with angle of insonation always <45°, as previously reported (Mancini et al., 2000).
Semen analysis, urine and seminal cultures and hormone evaluation
All patients underwent, after the ultrasound session, semen analysis, performed according to the WHO criteria. Leucocytospermia was defined as leucocytes in seminal plasma ≥1 × 106/mL (WHO, 2010). In addition, urine and seminal cultures were routinely assessed in all men.
Blood samples were drawn in the morning, after an overnight fast, for determination of blood glucose (by glucose oxidase method; Aeroset Abbott, Rome, Italy), HDL cholesterol and triglycerides (by automated enzymatic colorimetric method; Aeroset Abbott), LH, FSH, PRL, TSH, total testosterone (TT) by electrochemiluminescent method (Modular Roche, Milan, Italy) and sex hormone binding globulin (SHBG) by modular E170 platform electrochemiluminescence immunoassay (Roche Diagnostics, Mannheim, Germany). Calculated free testosterone was derived according to Vermeulen's formula (available at http: //www.issam.ch/freetesto.htm) (Vermeulen et al., 1999). Biochemical hypogonadism has been defined for testosterone levels <12 nmol/L (Wang et al., 2009). Secondary or primary hypogonadism has been defined for LH ≤ 9.4 or > 9.4 U/L, respectively, according to Tajar et al. (2010).
Identification of case patients and controls
The MetS was defined as described above. Subjects with three or ≥4 MetS components (n = 15 and n = 12 respectively) were compared with controls selected from the same cohort with a 1 : 3 ratio (n = 45 and n = 36 respectively). For each case, the first three following patients within the same series with the same age (± 4 years), body mass index (± 2 kg/m2), TT levels (± 4 nmol/L), smoking status (current/non-smoker), alcohol consumption (current/no consumption), past or present cryptorchidism (presence/absence), leucocytospermia (presence/absence), current positive urine and/or semen cultures (presence/absence) were taken as a control. Associations with a p value <0.05 were considered as significant.
Data were expressed as mean ± standard deviation (SD) when normally distributed, as median (quartiles) for parameters with non-normal distribution, and as percentages when categorical. Correlations were assessed using Spearman's or Pearson's method whenever appropriate. Differences between more than two groups were assessed with one-way anova or Kruskal–Wallis test, whenever appropriate. Unpaired two-sided Student's t-tests were used for comparisons of means of normally distributed parameters. In all other cases, Mann–Whitney U-test was used for comparisons between groups. Relative risk and 95% confidence interval were calculated for association of categorical parameters, and chi-squared test was used for comparisons. Stepwise multiple linear or logistic regressions were applied for multivariate analysis, whenever appropriate. All statistical analysis was performed on spss (Statistical Package for the Social Sciences, Chicago, IL, USA) for Windows 17.0.
Among the patients selected (mean age 36.0 ± 8.0), 178 (50.7%) had no MetS components, 99 (28.2%) one, 47 (13.4%) two, 15 (4.3%) three, 8 (2.3%) four and 4 (1.1%) five components. Overall, 27 (7.7%) satisfied MetS criteria, according to the IDF & AHA/NHLBI criteria. Subjects with MetS were older (43.7 ± 10.4 vs. 35.5 ± 7.4; p < 0.0001) and had significantly lower levels of TT (13.8 ± 6.5 vs. 16.7 ± 6.2; p < 0.05) compared to the rest of the sample. No difference in the percentage of subjects with current smoking was found comparing MetS and no-MetS subjects (18.5% vs. 28.5%, p = 0.178). In addition, no difference in the frequency of those consuming alcohol was observed between the two groups (22.2% vs. 25.6%, for MetS and no-MetS subjects respectively; p = 0.671). Subjects with diabetes mellitus were three (1.1%), representing 11.1% of the MetS group. The percentage of subjects under specific medications for MetS components (hypoglycaemic, n = 3; lower-lipid, n = 6; antihypertensive, n = 14) in the whole sample and in the MetS group is reported in Table 1.
Erectile dysfunction, assessed by the IIEF-15-EFD score, was complained of by 69 (19.9%) patients, in particular by 46.2 and 16.7% of the MetS and no-MetS subjects respectively. Premature ejaculation (PE), assessed by the PEDT score, was observed in 53 (15%) patients, particularly in 22.2 and 14.4% of the MetS and no-MetS subjects respectively.
Etiological factors involved in male infertility: classification, prevalence and associations with MetS
Table 2 shows a patient stratification according to etiological factors involved in male infertility (as in Krausz, 2011), and the prevalence of MetS for each infertility factor. Possible associations between MetS and different causes of infertility have been investigated. A higher prevalence of secondary hypogonadism was found in subjects with MetS compared to the rest of the sample (51.9% vs. 17.6%, respectively; p < 0.0001). In contrast, a lower frequency of varicocoele was observed in MetS subjects, according to both clinical and CDU criteria (Table 2). In particular, the presence of severe CDU varicocoele was negatively associated with MetS (RR = 0.28 [0.08–0.97], p < 0.05). No further association with other clinical abnormalities was observed (Table 2). In particular, no statistically significant difference in the prevalence of past or present cryptorchidism, leucocytospermia or current positive urine and/or semen cultures was observed comparing subjects with and without MetS (Table 2).
Table 2. Etiological factors involved in male infertility in the whole sample and in subjects with or without metabolic syndrome.% of subjects with MetS for each infertility factor
Etiological factors involvedin male infertility
% of the cohort (n = 351)
Without MetS n = 324
With MetS n = 27
% of subjects with MetS for eachetiological factor
Data are expressed as percentage.
CDU: colour-Doppler ultrasound.
aBiochemical hypogonadism has been defined for total testosterone levels < 12 nmol/L, according to Wang et al. (2009). Secondary or primary hypogonadism has been defined for LH ≤ 9.4 or > 9.4 U/L, respectively, according to Tajar et al. (2010).
bClinical varicocoele: any degree of varicocoele (Grade 1–3) according to Dubin & Amelar classification (Dubin & Amelar, 1970).
cCDU defined severe varicocoele = basal venous reflux increasing after Valsalva's manoeuvre at CDU (according to Isidori & Lenzi, 2008; Lotti et al., 2009).
A statistical analysis comparing patients with and without MetS has been performed. *p < 0.05; **p < 0.0001.
In the same Table, the prevalence of MetS subjects under each clinical condition involved in male infertility is also reported. It is evident that the secondary hypogonadism is the most common MetS-associated condition.
Ultrasonography testis parameters
At univariate analysis, both testis and epididymal inhomogeneity were associated with MetS components (RR = 1.43 [1.16–1.76], p = 0.001 and 1.27 [1.03–1.57], p < 0.05, respectively, for each increment in the number of MetS components). In an age-adjusted logistic modelling, only testis inhomogeneity retained a significant association, and the HR for testis inhomogeneity as a function of MetS components was 1.36 [1.09–1.70], p < 0.01 (for each increment in the number of MetS components, see Figs 1A and 2A). These results were confirmed even after exclusion of subjects with diabetes mellitus (n = 3) from the analysis (data not shown). Figure 2A shows associations between individual MetS components and testis inhomogeneity. After adjusting for age, the only factor significantly associated with testis inhomogeneity was increased waist (HR = 2.25 [1.23–4.09], p < 0.01).
No further association between MetS and other testicular ultrasound parameters, including volume, was observed (data not shown).
Hormonal and seminal parameters
Figure 1B, shows that TT declines as a function of MetS components (p for trend at anova < 0.0001). This result was confirmed even after exclusion of subjects with diabetes mellitus (n = 3) from the analysis (data not shown). Differences in TT levels between those having two or three MetS components were not statistically significant (p = 0.263; Figure 1B).
Table 3 reports associations between MetS and hormonal parameters before and after adjusting for age. In an age-adjusted model, both TT and also calculated free T were negatively associated with MetS components, whereas gonadotropins, TSH and PRL were not (Table 3). These results were confirmed even after exclusion of subjects with diabetes mellitus (n = 3) from the statistical analysis (data not shown). By using an age-adjusted, iterative logistic modelling, we calculated the HR for hypogonadism (TT < 12 nmol/L) as a function of MetS components, taken together or individually (Fig. 2B). Factors significantly associated with hypogonadism were increased waist (HR = 3.1 [1.6–5.9], p = 0.001), low HDL cholesterol (HR = 2.4 [1.16–4.96], p < 0.02), and high triglycerides (HR = 2.4 [1.12–5.1], p < 0.05). We also found that testis inhomogeneity was a risk factor for hypogonadism, even after adjusting for age (HR = 2.99 [1.71–5.24], p < 0.0001). We therefore introduced TT levels in the logistic model describing the association between MetS factors and testis inhomogeneity, as previously described (see before). Adding TT as a further covariate, besides age, significantly attenuated the relationship between number of MetS components and testis inhomogeneity (HR = 1.24 [0.98–1.56], p = 0.07).
Table 3. Univariate and multivariate (age-adjusted) associations between metabolic syndrome components and hormonal parameters
Table 1 shows semen parameters of the whole sample and of subjects with or without MetS. As sexual abstinence had a wide variation in our population, eventually affecting semen parameters, the possible association between sexual abstinence and semen parameters has been investigated by univariate analysis, finding no correlation either in the whole sample or in MetS-stratified groups (data not shown). At univariate analysis, progressive motility and normal morphology were negatively related to number of MetS components (both p < 0.0001 at anova and Kruskal–Wallis respectively). As diabetes mellitus or current male genital infections may affect semen parameters, statistical analyses have been performed excluding iteratively subjects with diabetes mellitus (n = 3), leucocytospermia (n = 29) or current positive urine and/or semen cultures (n = 25). After the exclusion of these subjects, previous associations were confirmed at anova for progressive motility and Kruskal–Wallis for normal morphology (data not shown). When age and TT were introduced as covariates in a multivariate model describing the relationships between MetS components and sperm parameters, only sperm morphology retained a significant association (B = −1.418 ± 0.42; p = 0.001) (see Fig. 1C). In addition, as smoking habit/alcohol consumption and use of medications might negatively affect sperm parameters, these confounders were iteratively introduced as further covariates in the aforementioned multivariate model. Neither smoking/drinking habit nor use of medications were associated with abnormal sperm morphology while MetS still resulted a statistically significant determinant in both models (B = −1.318 ± 0.45; p = 0.003 and B = −1.333 ± 0.24; p < 0.005 respectively).
Figure 2C, reports, as a Forest Plot the HR for abnormal morphology (<4%, WHO, 2010) as a function of individual MetS components and MetS overall. In an age- and testosterone-adjusted logistic model, only hypertension was associated with abnormal morphology (HR = 1.99 [1.11–3.54], p = 0.02). The percentage of normal sperm morphology (≥4%, WHO, 2010) in subjects with or without IDF & AHA/NHLBI-defined hypertension was 3.0 [1.0–7.0] vs. 6.0 [2.0–10.0] (p < 0.01). As testis inhomogeneity was an important risk factor for abnormal morphology, even in an age- and TT-adjusted logistic model (HR = 3.68 [1.99–6.82], p < 0.0001), we tested the effect of introduction of testis inhomogeneity on the relationship between IDF & AHA/NHLBI-defined hypertension and abnormal morphology. In an age- and TT-adjusted logistic model, both testis inhomogeneity and hypertension were found to be independent risk factors for poor sperm morphology (HR = 3.44 [1.82–6.53], p < 0.0001 and HR = 1.86 [1.02–3.39], p < 0.05 respectively). No other associations between MetS and semen parameters were observed (data not shown).
When subjects with ED (IIEF-15-EFD score <26) were divided according to the presence of MetS, those satisfying MetS criteria had a higher ED prevalence (46.2% vs. 16.7%, p = 0.001). These results were confirmed after exclusion of subjects with diabetes mellitus (n = 3) from the analysis (data not shown). The association was confirmed in a logistic model, after adjustment for age and TT (HR = 2.86[1.18–7.31], p < 0.05). At univariate analysis, the risk of ED increased as a function of MetS components (RR = 1.56 [1.22–2.01] p < 0.0001 for each increment in the number of MetS components), and this association was confirmed after adjusting for the aforementioned confounders (HR = 1.45[1.08–1.95], p < 0.02, for each increment in the number of MetS components, Fig. 1D). In an age- and TT-adjusted logistic model, the MetS factors positively associated with ED were increased waist (HR = 2.92 [1.28–6.63], p < 0.02) and blood pressure (HR = 2.33 [1.1–4.9], p < 0.05, Fig. 2D).
When penile basal CDU was considered, arterial peak systolic velocity and acceleration were negatively related to MetS components, even after adjusting for age and TT (B = −0.748 ± 0.314, p < 0.02 and B = −0.201 ± 0.092, respectively, both p < 0.05). These results were confirmed after exclusion of subjects with diabetes mellitus (n = 3) from the analysis (data not shown).
No association between PEDT score and MetS was observed (data not shown) and no difference in PE prevalence was found comparing subjects with or without MetS (22.2% vs. 14.4%, p = 0.283).
At univariate analysis, patients with MetS showed a significantly higher score in MHQ-somatization (MHQ-S; 2.962 ± 2.821 vs. 1.857 ± 2.266, p = 0.022) and depression scales (MHQ-D; 4.578 ± 2.845 vs. 3.004 ± 2.516, p = 0.011). Scores for these two MHQ subscales were associated with numbers of MetS factors in both unadjusted and in age- and TT- adjusted models (MHQ-S: unadjusted r = 0.126, adjusted B = 0.66 ± 0.03, for each increment in the number of MetS components, both p < 0.05; MHQ-D unadjusted r = 0.197, p = 0.001, adjusted B = 0.69 ± 0.03 for each increment in the number of MetS components, p < 0.02).
The correlations between MetS and seminal, ultrasound and sexual parameters resulting statistically significant were further assessed by comparing subjects with MetS with a 1 : 3 ratio matched controls (matched for age, body mass index, TT, smoking habit, alcohol consumption, past or present cryptorchidism, leucocytospermia, current positive urine and/or semen cultures; see Table 4). Even in the case–control analysis, subjects with three or more MetS components showed a significant increase in prevalence of abnormal morphology and testis inhomogeneity (Table 4). In addition, subjects with ≥4 MetS factors had more often ED (Table 4).
Table 4. Comparisons between subjects with metabolic syndrome (MetS) and 1 : 3 ratio matched controls (matched for age, body mass index, total testosterone, smoking habit, alcohol consumption, past or present cryptorchidism, leucocytospermia, current positive urine and/or semen cultures). Comparison between subjects with ≥4 MetS components and three MetS components
Case patients (with 3 MetS factors, n = 15)
Controls (matched 1:3, n = 45)
Case patients(with ≥4 MetS factors, n = 12)
Controls(matched 1:3, n = 36)
Patients with≥4 vs. 3 MetS factors, p
Data were expressed as mean ± standard deviation when normally distributed, median (quartiles) when not normally distributed, and as percentages when categorical.
BMI: body mass index, ED: erectile dysfunction, IIEF-15-EFD: International Index of Sexual Function-15 erectile function domain. Associations with a p-value <0.05 were considered as significant, and reported in bold.
39.9 ± 8.5
39.3 ± 7.2
48.5 ± 10.9
41.5 ± 10.1
Total testosterone (nmol/L)
16.1 ± 6.6
16.8 ± 5.4
10.8 ± 5.0
13.5 ± 4.8
29.5 ± 4.7
27.6 ± 4.6
34.5 ± 6.0
30.8 ± 5.0
Current smoker, %
Current alcohol consumption, %
Past or present cryptorchidism
Current positive urine and/or semen cultures, %
Sperm concentration, × 106/mL
Spermatozoa per ejaculate, × 106/mL
Sperm progressive motility, %
42.1 ± 19.2
42.3 ± 18.6
35.0 ± 13.9
36.9 ± 17.8
Sperm morphology, % normal forms
Testis inhomogeneityat ultrasound, %
ED prevalence(IIEF-15-EFD < 26) , %
Effect of increasing number of MetS factors on reproductive health
To assess if, in MetS subjects, having ≥4 components has a more negative impact on reproductive health than having only three components, a comparison of the aforementioned parameters was performed between the two groups (Table 4). Subjects with ≥4 MetS factors were older, had a higher body mass index, were more often affected by ED and characterized by a lower TT levels compared to those with three components only (Table 4). No significant difference in normal sperm morphology and prevalence of testis ultrasound homogeneity was observed between the two groups (Table 4).
This study demonstrates, for the first time, a component-dependent, stepwise association between the presence of MetS and abnormal semen parameters, along with several psychobiological factors which might further negatively impact male reproductive potential. In particular, we confirm, in subjects seeking medical care for male infertility, a clear-cut association between increasing numbers of MetS components and severity of male hypogonadism, previously reported in other patient populations (Muller et al., 2005; Corona et al., 2006, 2009a). Both total and calculated free T were reduced as a function of MetS factors. In particular, we found that having ≥4 MetS components exert a much more negative impact on TT levels than having only three. The finding that gonadotropins did not rise according to the androgen fall, indicates an hypogonadotropic nature of the hypogonadism, as previously described in studies performed on animal models of MetS (Filippi et al., 2009; Corona et al., 2011b) and in human cohorts (Corona et al., 2009b; Dandona & Dhindsa, 2011). Accordingly, MetS triplicates the risk of having secondary hypogonadism, this is the most common, infertility-associated condition in subjects with MetS. When the contribution of individual MetS components was investigated, we found that increased visceral adiposity (waist) and dyslipidaemia (reduced HDL cholesterol and elevated triglycerides) were the main determinants of testosterone decline. Interestingly, increased waist was found as the main determinant for testis inhomogeneity, indicating a specific pathogenetic role for adiposity-related factors in testicular abnormalities. Testis inhomogeneity at ultrasonography is suggestive of atrophy and fibrosis (Cohn et al., 1996; Loberant et al., 2010) and may reflect changes in the functional activity of the testis (Lenz et al., 1993; Behre et al., 1995). In fact, in the ageing testis, atrophy of the tubular elements and proliferation of interstitial ones produce an exaggeration of the normally unapparent septal and interstitial architecture, resulting in a striated appearance (Loberant et al., 2010). While a similar appearance is often observed in the elderly and considered normal, in young subjects it is recognized in a variety of pathological conditions, including hypogonadism (Cohn et al., 1996; Loberant et al., 2010; Migaleddu et al., 2012). Present data suggest that a secondary hypogonadism is predominantly associated with MetS, and in particular to visceral adiposity. Visceral adiposity is also the main determinant of testis inhomogeneity. The latter can derive from a decreased testicular stimulation, because of a central defect, and/or from a direct effect of visceral adiposity on testis itself (mixed hypogonadism), as previously reported in other patient cohorts (Corona et al., 2008b, 2009b). Several molecules, such as insulin, oestrogens, leptin, tumour necrosis factor α (TNFα) or other adipokines, mainly related to increased adiposity, have been hypothesized to induce MetS-related hypogonadism, acting both at a central or at a peripheral level (see for review Corona et al., 2011b). In contrast with the positive association between hypogonadism and MetS, varicocoele was less often found in subjects fulfilling MetS criteria, thus confirming the view that obesity is a protective factor against varicocoele (Handel et al., 2006; Nielsen et al., 2006; Tsao et al., 2009).
We here describe that there is a specific association between MetS and some sperm parameter abnormalities, including abnormal forward motility and morphology. These are the two well recognized sperm parameters for successful fertilization both in vitro and in vivo (Chan et al., 1989; Joshi et al., 1996; Donnelly et al., 1998). However, after adjusting for testosterone levels, only abnormal morphology retained a specific association with MetS, suggesting that hypogonadism, more than MetS itself, is responsible for the decreased progressive motility. We also reported that testis inhomogeneity was associated with abnormal morphology, as previously described (Lenz et al., 1993). The MetS factor specifically associated with this abnormality was hypertension, and not visceral adiposity or other metabolic disturbances (hyperglycaemia, dyslipidaemia). Hypertension increased by a factor of two, the risk of abnormal morphology, independently from other factors, such as testis inhomogeneity and testosterone levels. Mechanisms by which hypertension might affect sperm morphology are far from being understood (Kasturi et al., 2008). Akagashi et al. (1997) found an extreme spermatogenic impairment in the form of atrophic seminiferous tubules devoid of spermatids in stroke-prone spontaneously hypertensive rats. Very recently, a study of semen quality in hypertensive men demonstrated an association between hypertension with more fragmented/abnormal sperm DNA and with increased expression of clusterin, an apoptosis-associated protein, which identifies a subset of morphologically altered spermatozoa not necessarily dead (Muciaccia et al., 2012). In that study, (Muciaccia et al., 2012) the link between elevated blood pressure and abnormal morphology/DNA fragmentation was not clarified, but it was hypothesized that hypertension induces, within the testis, a generally altered vascular status characterized by enhanced ROS generation and limited antioxidant defence (Sharma et al., 1999; Brownlee, 2005; Amaral et al., 2008). Another recently published report (Mbah et al., 2012) indicates that treating normotensive, oligospermic men with a low-dose angiotensin converting enzyme inhibitor (lisinopril, 2.5 md/daily) ameliorates, in a rigorous placebo-controlled, crossover design, semen parameters, including sperm number, motility and morphology. In a rabbit model of MetS, also characterized by hypertension, we observed an increase in advanced glycation end products and their receptor (RAGE) in the seminiferous epithelium and in the cytoplasm of Sertoli cells, spermatocytes and spermatids respectively (Mallidis et al., 2011). Accumulation of advanced glycation end products and their receptor RAGE are thought to be capable of generating, promoting and/or amplifying oxidative stress and its detrimental consequences, including DNA damage (Sakkas & Alvarez, 2010). More studies are advisable to verify whether hypertension is associated with DNA fragmentation in subjects with MetS.
Erectile dysfunction (IIEF-15-EFD score <26) was present in ~20% of the study cohort. This prevalence is essentially in agreement with previous reports in infertile men (Jain et al., 2000; O'Brien et al., 2005; Shindel et al., 2008; Lotti et al., 2012), but is at least double than that observed in the general Italian population, when subjects with a similar mean age are considered (Parazzini et al., 2000; Mirone et al., 2004). The prevalence of ED in our patient population was further increased by the MetS condition, with hypertension and visceral adiposity as the main determinants. In addition, we found that having ≥4 MetS components exerts the maximal negative impact on sexual function. Also PCDU-measured penile blood flow was decreased as a function of MetS severity. Similar results were reported in other studies (Corona et al., 2006; Demir et al., 2009; Tomada et al., 2011). Interestingly, in this study, the association between an increasing number of MetS components and both ED and reduced penile blood flow were independent from low testosterone. The lack of association between MetS and premature ejaculation (PE or PEDT score) was expected according to previous studies of our group (Corona et al., 2008c, 2011c, 2012), where hypogonadism was more prevalent in delayed than in premature ejaculation.
Not only ED but also depressive symptoms and dysfunctional sexual relationship have been previously reported in subjects with couple infertility (Shindel et al., 2008; Smith et al., 2009; Lotti et al., 2012). It is well known that infertile men may develop feelings of inadequacy, anxiety, guilt and depression (Irvine & Cawood, 1996; Seidman & Roose, 2000). We now report that MetS severity further increases these negative emotions. In fact, both depressive symptoms and somatized anxiety were more prevalent in infertile men with MetS. MetS is highly prevalent among patients with a history of depression, especially those with current major depression (Heiskanen et al., 2006), perhaps because of increased body fatness or lifestyle alterations. However, previous prospective data indicated that depression, hostility and anger predict increased risk of MetS (Goldbacher & Matthews, 2007). Association between MetS and somatized anxiety was previously described by our group in subjects with sexual dysfunction (Corona et al., 2006). Somatization, which is the expression of physical symptoms in the absence of medically explained physical illness, might be viewed as a part of maladaptive thought or behaviour, or just as a dysfunctional response, to the infertility condition itself – an often unaccepted or unrecognized medical condition. We here report that having MetS increases somatization and depression scoring, thus increasing the high attention devoted by these subjects to their body and possible bodily malfunctioning and therefore freezing several aspect of couple's sexual behaviour, including sexual desire and intercourse frequency (Corona et al., 2008d), further exacerbating the infertility condition. A large body of evidence indicates that infertility evokes more general psychological distress in women than in men (Jordan & Revenson, 1999). Accordingly, supportive psychosocial interventions are thought to be more beneficial for the female than for the male partner of the infertile couple (Hämmerli et al., 2009). It is possible that MetS-associated negative emotions and sexual dysfunctions reduce this gender-related gap in psychological distress. Further studies are advisable to clarify this point.
In conclusion, MetS is a set of metabolic risk factors originally proposed as a diagnostic tool because of its potential in identifying subjects at risk for diabetes and cardiovascular diseases, with the idea that its components could somehow have a synergistic effect on predicting unfavourable events. Later on, it became clear that other medical conditions are comorbid with MetS, including male hypogonadism and erectile dysfunction (Corona et al., 2009a, 2011a,b). We now report that an increasing number of MetS factors are dose-dependently associated with relevant organic (poor sperm quality, hypogonadism, ED) and psychological (depression, somatization) features that might affect reproductive outcomes of men seeking medical care for couple infertility. This might tailor ad hoc therapeutic intervention. Behavioural interventions targeting lifestyle factors, such as dietary practice and physical activity, might ameliorate not only metabolic and psychological parameters but also male infertility, as has been demonstrated for female infertility. Recognizing the umbrella condition termed ‘metabolic syndrome’ might therefore represent a unique occasion to improve not only reproductive health but also psychological, sexual and overall health.