Obesity and Male Reproductive Potential

Obesity is a well recognized risk factor for female infertility (Pasquali et al, 2003). However, its relation to decreased sperm count was not documented until recently (Jensen et al, 2004; Magnusdottir et al, 2005). It is believed that with the increasing prevalence of sedentary life styles and dietary changes, obesity is emerging as an important cause of adverse health outcomes, including male infertility. Male factors alone constitute 25%–30% of all cases of infertility, and they contribute to another 30% in combination with female factors. Known etiologies of male infertility include cryptorchidism, testicular torsion or trauma, varicocele, seminal tract infections, anti-sperm antibodies, hypogonadotropic hypogonadism, gonadal dysgenesis, and obstruction of the reproductive channels (Oehninger, 2000). Obesity was recently proposed for addition to this list (Jensen et al, 2004). Thus, obesity is a chronic disease resulting in compromised physical and psychological well-being, and it may contribute to reduced fecundity as well. It is known that the incidence of obesity is reaching epidemic levels in the western world. For example, in the United States, the incidence of obesity increased from 12% to 17.9% between 1991 and 1998. In the same period, the incidence of male obesity increased from 11.7% to 17.9% (Mokdad et al, 1999). Recently, the prevalence of male obesity in the US was reported to be 30.6% (Hedley et al, 2004). It is possible that the increasing prevalence of overweight and obesity accounts for a portion of the trend, albeit a widely debated one, of decreasing sperm counts over recent decades. Although complicated by varying sample sizes and methodologies for the assembled data, it has been estimated that sperm counts have been decreasing by as much as 1.5% each year in the United States, a finding similar to those for other Western countries and not present in other regions where obesity is less prevalent (Swan et al, 2000). These findings suggest a possible link between life style changes, obesity, semen quality, and possibly male fertility (Jensen et al, 2004). Obesity in adolescent girls is associated with increased height and earlier pubertal development (Dunger et al, 2005). There is some evidence that development of sexual characteristics takes place earlier as well (Adair and Gordon-Larsen, 2001). In males, the relationship of body weight with sexual development seems to be reversed. Wang et al studied a sample of 1501 boys and 1520 girls aged 8 to 14 years. They found that the prevalence of obesity was lower in early-maturing boys and higher in early-maturing girls, compared with the group who had average or late sexual maturity (Wang, 2002). This finding was challenged later by Laron, who found no significant difference in the indicators of sexual maturation between obese and nonobese adolescents in both genders; however, the number of patients in this study was small (Laron, 2004). On the other hand, pubertal development was shown to negatively influence the secretion of adiponectin by adipose tissue. Adiponectin has mainly antidiabetic and antiatherogenic effects. In a multiple regression model adiponectin levels in males were shown to be negatively correlated to the stage of pubertal development (r2 = .206, P < .001), body mass index (BMI) (r2 = .034, P = .021), and testosterone levels (r2 = .039). This analysis demonstrated that the decrease in adiponectin levels throughout pubertal development stages is more pronounced in obese males than controls (Bottner et al, 2004). Obese males usually express a characteristic hormonal profile described as “hyperestrogenic hypogonadotropic hypogonadism.” In fact, both total and free blood testosterone levels are shown to be decreased in obese men. Total body fat, intra-abdominal fat, and subcutaneous fat have all been associated with low levels of total and free testosterone (Strain et al, 1982; Haffner et al, 1993; Tsai et al, 2004). It appears that central obesity, in particular, is associated with a decrease in circulating androgen levels. The decrease in androgen levels is proportional to the degree of obesity (Giagulli et al, 1994). In obese males, adrenal androgens are diminished as well (Tchernof et al, 1995). The origin of hypoandrogenism in obese males is multifactorial (Table). It is primarily attributable to an increase in circulating estrogens that appear to result in relative hypogonadotropism, although the diminished levels of sex hormone-binding globulin (SHBG) in obese individuals will by itself result in reduced total testosterone levels, even in the face of unchanged production. In fact, weight correlates negatively with blood testosterone levels and testosterone/estradiol ratio (Fejes et al, 2006). Both estrone and estradiol are increased in obese males compared to controls (Schneider et al, 1979). The aromatization of C19 androgens like testosterone and androstenedione is a key step in estrogen biosynthesis and is catalyzed by the aromatase enzyme, a product of the CYP19 gene. It is believed that the increase in estrogens in obese males is due to increased conversion of adrenal and testicular androgens owing to the increase in available aromatase enzyme in the fatty tissue (de Boer et al, 2005). Estrogen production by adipose tissue is dependent on the availability of androgenic precursors in the circulation (Simpson et al, 1999). Estrogens are biologically active at much lower concentrations and production rates than androgens (testosterone production in adult males is several mg per day, whereas estradiol production in women averages 100 μg per day); thus small increases in aromatization of androgens can result in changes in estrogen levels that are substantial in relation to the range of normal circulating levels. Increased circulating estrogen levels resulting from increased peripheral conversion among the obese may therefore inappropriately enhance negative feedback on gonadotropin secretion. Estradiol has significant biologic activity in the hypothalamus and the pituitary. Estrogen receptors (ERα and ERβ) are present in several hypothalamic nuclei and in pituitary gonadotropes, indicating that estrogen modulates the activity of hypothalamus-pituitary axis. Estrogen acts on the hypothalamus to affect gonadotropin-releasing hormone (GnRH) pulses and at the pituitary level to regulate gonadotropin (follicle-stimulating hormone [FSH] and luteinizing hormone [LH]) secretion (Akingbemi, 2005). In severe obesity, pituitary gonadotropin secretion appears suppressed with normal or decreased levels of LH in the presence of decreased levels of testosterone (Giagulli et al, 1994). Because aromatization also occurs in the hypothalamus itself, precise dissection of the contribution of increases in circulating estradiol associated with obesity to overall negative steroidal feedback on pituitary gonadotropin secretion is problematic. The hypothesis that elevated estrogen plays an important role in the androgen abnormalities in obesity is strongly supported by observations of the effect of the aromatase inhibitor letrozole on obese men. Among 10 severely obese men with markedly low testosterone levels and clinical symptoms of hypogonadism, letrozole administration for 6 weeks increased testosterone levels more than 3-fold from a mean baseline of 7.5 ± 1 nmol/l (mean ± SE) while diminishing circulating estradiol from 120 ± 20 to 70 ± 9 pmol/l (mean ± SE). Changes in levels in both steroids were highly statistically significant (P < .001), and were not accompanied by changes in SHBG. A 3-fold statistically significant increase in both LH and FSH occurred during treatment as well (de Boer et al, 2005). Unfortunately, this study did not include data on measures of semen quality. Raman and Schlegel reported a tripling of the testosterone-estradiol ratio among 16 overweight infertile men treated with the aromatase inhibitor anastrozole but did not report changes in semen parameters for this subgroup of a study cohort that included men with other known causes for male infertility, including varicocele and Klinfelter syndrome (Raman and Schlegel, 2002). The results of this suggest also that aromatase inhibition increases in testosterone levels and decreases in estradiol in nonobese males, such that the “correction” of hypoandrogenism in obese males during aromatase inhibition is an extension of a general effect of such medications. To the extent that lower circulating testosterone levels reflect decreased testicular testosterone production, lower intratesticular testosterone levels would be expected as well. Insulin resistance, a predisposition of obesity, has also been reported to be associated with low testosterone levels. Age-adjusted fasting insulin and C-peptide were shown to be inversely correlated to total and free testosterone in men (Tsai et al, 2004). This association is confounded by the independent relation between SHBG and insulin resistance (Stellato et al, 2000). However, after adjusting for SHBG levels, low testosterone levels remain correlated with insulin resistance (Tsai et al, 2004). The direction of this association is not clear. In a recent review, men with type 2 diabetes had a mean decrease of testosterone levels of (–71.5 ng/dl, 95% CI: −116.4 to −26.8 ng/dl) compared to controls. Moreover, the analysis of prospective studies showed that the risk of developing type 2 diabetes was 42% (RR = 0.58, 95% CI: 0.39 to 0.87) lower in men with high testosterone levels (range, 449.6–605.2 ng/dl) than men with low testosterone levels (range 213.2–446.7 ng/dl) (Ding et al, 2006). Whether the association between glucose intolerance and low androgen levels is an independent relation remains a matter of debate, as any association between hyperinsulism and hypoandrogenism is confounded by the association of obesity with hyperinsulism and insulin resistance. The third suggested cause of hypoandrogenism in obese males is sleep apnea. Patients with sleep apnea often have fragmented sleep course due to repetitive episodes of upper airway obstructions and hypoxia followed by arousal (Young et al, 2004). It has been demonstrated that patients with fragmented sleep have a blunted nocturnal rise of testosterone (Luboshitzky et al, 2001). Patients with obstructive sleep apnea have lower mean testosterone and LH values compared to both young and middle-aged controls. Morning testosterone levels were also found to be lower in patients with obstructive sleep apnea (Luboshitzky et al, 2005). Finally, in patients with obstructive sleep apnea, weight loss increases testosterone levels (Semple et al, 1984). These alterations in testosterone levels are likely to contribute to hypogonadism in a number of obese males. This observation is confounded by obesity, such that an independent effect on testosterone levels of sleep apnea separate from obesity requires further confirmation. It is mandatory to mention leptin in the context of obesity and infertility. The role of leptin in reproduction was discovered in ob−/ob− mice. The lack of obesity (ob) gene encoding for leptin caused obesity and infertility in both male and female mice. Leptin administration in leptin deficient ob−/ob− mice restored fertility (Chehab et al, 1996). Obesity is associated with increased levels of circulating leptin, since white adipose tissue is the main site of leptin synthesis (Margetic et al, 2002). The effect of excess leptin on male reproduction is unknown. In hypogonadal men, testosterone substitution normalized elevated serum leptin levels (Jockenhovel et al, 1997). Leptin effects on reproduction are largely thought to be central, altering the hypothalamic pituitary axis (Margetic et al, 2002). Recent discovery of leptin in semen suggests a peripheral function as well (Aquila et al, 2005). Body mass index is associated with alterations in sperm parameters in several reports. In a recent study investigating factors associated with semen quality among couples who visited an assisted reproduction clinic, the prevalence of obesity among men with infertility (defined as at least 2 findings among density less than 10 million, total count less than 20 million, motility less than 30%) was 3 times greater than among male partners of couples with idiopathic or female factor infertility. Groups were small in this study, and specific etiologies for lower sperm parameters among obese subjects were not described. Among the 47 men in this study with normal semen parameters at enrollment, but not among men with male infertility, sperm density and total count showed a statistically significant negative association with increasing BMI (Magnusdottir et al, 2005). Among a sample of 1558 Danish military recruits, several semen parameters were correlated with body mass index, with less favorable values reflecting spermatogenesis observed in individuals with either above or below the BMI range of 20–25 kg/m2 (Jensen et al, 2004). After correction for multiple factors, including exposures, illnesses, and abstinence period prior to sample collection, sperm concentration and total count per ejaculate among men with BMI. 25 kg/m2 were reduced by 26.1% and 23.9%, respectively, when compared to the reference group with BMI between 20 and 25 kg/m2. The frequency of sperm density less than 20 million/ml was 29% among the overweight men compared to 21.7% in the normal weight reference group. Reductions in these measures were similar among the subjects with BMI less than 20 kg/m2, such that total sperm count and sperm concentration exhibited an inverted U-shaped distribution relative to BMI. The endocrine findings in the heavy and thin men trended in opposite directions relative to the reference BMI group in several instances, such that total testosterone, FSH, inhibin B, and SHBG were lower and estradiol higher in the overweight group relative to the reference group, while the reverse of these observations obtained for the leaner men. Thus, putative endocrine mechanisms underlying the apparent reduction in spermatogenesis, if any, among the lean and overweight subgroups can be inferred to be quite different. In a study of normozoospermic partners in an infertile population, sperm concentration was reduced among men with BMI greater than 30 compared to leaner members of the study group (Koloszar et al, 2005). Kort et al described the relation between sperm parameters and BMI in a generally overweight selection of subjects. An index of semen quality appeared to exhibit a decline with increasing BMI, as did the number of normal sperm per ejaculate. In his study, the total number of normal spermatozoa correlated negatively with body mass index in 520 semen analyses. The total number of normal motile spermatozoa differed statistically according to weight group (BMI < 25 kg/m2 = 18.6 × 106; BMI ≥ 25 kg/m2 and < 30 kg/m2 = 3.6 × 106; and BMI ≥ 30 = 0.7 × 106, P < .05, ANOVA) (Kort et al, 2005). The relationship between BMI and sperm parameters was not preserved when obesity was expressed as waist/hip ratio. An absence of correlation with semen parameters with waist/hip ratio has been reported elsewhere (Fejes et al, 2005). A recent report showed that bariatric surgery may result in male infertility. Six severely obese men with at least 1 previous child were found to be azoospermic after Roux-en-Y gastric bypass operation (di Frega et al, 2004). To the extent that obesity is causal of diminished sperm count and sperm density, such findings may relate to endocrine abnormalities of obese males. Where lower total and free levels of testosterone are found, it can be inferred (though it has not been shown) that these levels reflect decreased testicular testosterone production, and therefore lower intratesticular testosterone levels would be expected as well. Intratesticular testosterone levels (normally 100-fold greater than circulating concentrations) are correlated with spermatogenesis (Coviello et al, 2004; Coviello et al, 2005; Jarow and Zirkin, 2005; Matthiesson et al, 2005). It is not known whether the modest reductions in testosterone levels associated with obesity are accompanied by a reduction in intratesticular testosterone concentrations sufficient to explain the reductions in sperm count that have been observed in obese subjects. Although the observations of Raman and Schlegel that effecting an increase in low testosterone-estradiol ratios with aromatase inhibition is accompanied by improvement in semen parameters during aromatase inhibition hints that this may be so, it is clear that this issue requires further investigation (Raman and Schlegel, 2002). Moreover, it is important to note evidence that diminished intratesticular testosterone concentrations alone, absent concomitant suppression of FSH, may be insufficient for effective suppression of spermatogenesis (Narula et al, 2002). The consensus observation that obesity and overweight are associated with a reduction in semen quality must be qualified by the observation that the studies cited here demonstrate only trends, and many overweight and obese subjects have consistently been shown to exhibit normal semen quality. Moreover, it is important to consider the possibility that nutritional or factors associated with overweight and obesity, than a effect of obesity per for these observed In it is that decreased sperm quality has not been found in men with higher (Jensen et al, 2004; Magnusdottir et al, 2005). However, is in the with for the hypothesis that alterations of sperm parameters associated with obesity can be to suppression of the axis by elevated estrogens from peripheral and resulting decreased testosterone production in low levels of circulating testosterone and intratesticular testosterone (Table). The role of estrogen in male reproductive health was with the that to with activity may reproductive health et al, 2001). Obese men have been shown to exhibit higher levels of circulating estradiol elevated ratios in multiple studies (Schneider et al, 2004; 2006). Estrogens may affect spermatogenesis the as well as by alterations in gonadotropin secretion by the pituitary. estrogen has a effect on spermatogenesis in et al, 2003). However, elevated circulating estradiol owing to increased peripheral aromatization in obesity may not affect intratesticular estrogen It is that elevated circulating estrogen levels, as among obese men, have not been to infertility. is an it may in the observation that obesity increases the and levels of and endocrine that can affect fertility (Magnusdottir et al, 2005). The evidence cited thus to the that obesity negatively sperm production and total further studies will be to whether this effect is a cause for infertility attributable to obesity by To the extent that reduced total sperm counts and sperm concentration result from obesity, it is also that fertility with total count and sperm It has been shown that spermatogenesis and sperm parameters are correlated to and infertility of a et al demonstrated that the ratio of for subjects × sperm compared to those × was CI: This study also demonstrated that factors in addition to sperm density, such as may be important to and in this evidence that obesity sperm is et al, 2002). In couples to for the the of increases with sperm count to et al of the alterations in the of obese males can affect reproductive and to or infertility. The association between obesity and can be shown to another by obesity may contribute to infertility. is as the or to or an sufficient for sexual In men symptoms of of the subjects are found to be overweight or obese et al, 2000). The relation between obesity and can be in by the elevated levels of several in obese These of are associated with that is to male the et al, 1999). It is not clear if the association is due to an independent effect or due to risk factors that are associated with obesity et al, 1999). In fact, risk factors for such as and have with et al, 2000). to the sexual found in obese males 2005). Whether obesity is with or risk factors or it is that obese men have a higher incidence of that sexual life and In a study, obese men who to a loss of or more in total body weight by and increasing level of physical activity had a higher of weight loss and improvement in than controls et al, 2004). It is known that is associated with infertility. In a recent study where the for of infertile men reported an compared to of the group et al, 2005). In a of health obesity was associated with a relative risk for et al, 2003). of the possibility that infertility found in an obese may relate to underlying or abnormalities that affect spermatogenesis of obesity is important both to and factors that contribute to obesity are often and on the of several in association with factors such as and activity levels and 2001). This multifactorial the of cases of However, are where obesity and infertility can result from a or These causes are and measures and mechanisms the effects of obesity itself on male infertility. of such couples to be the to abnormalities to Klinfelter syndrome is the disease infertility in men. This syndrome results from a in sex The is The is by a analysis of blood 2003). The in this syndrome are due to during in development or less in patients can be obese with of fat distribution and however, obesity is not a They have small hypogonadism, and severe in spermatogenesis resulting in severe or and 1999). In adult of decreased and It is not known if the in the from an in or is due to lack of from the 2004). syndrome is by diminished hypogonadotropic hypogonadism, and It is the cause of obesity 2005). This is an disease caused by the absence of the either or loss of the with presence of 2 can of that for of cases 2004). is found in of patients with this syndrome et al, 2002). and spermatogenesis at the late level and 1999). resulting in male obesity and infertility are the and These are The between these 2 is not For many they constitute of a et al, 2000). They include multiple including hypogonadism, and obesity et al, 2003). is found mainly in Known abnormalities include in the to is but some patients have normal testicular function and spermatogenesis and 1999). The in obesity and sperm count suggests a link between obesity and male Obesity is associated with spermatogenesis and The spermatogenesis is mainly due to hypoandrogenism and the effect of increased levels of these factors can affect the of a male to in the of a child This is a for more studies to the extent of the correlation between obesity and decreased sperm count and to the contribution of obesity to male infertility. In the absence of clinical treatment for obese infertile males are and be the hormonal profile and the