- Email: [email protected]
High body mass index has a deleterious effect on semen parameters except morphology: results from a large cohort study
High body mass index has a deleterious effect on semen parameters except morphology: results from a large cohort study phanie Belloc, Pharm.D.,a Martine Cohen-Bacrie, M.D.,a Edouard Amar, M.D.,b Vincent Izard, M.D.,c Ste ac, M.D.,a and Jacques de Mouzon, M.D., M.P.H.e Moncef Benkhalifa, Ph.D.,d Alain Dalle a rest, Neuilly sur Seine; d Service de Laboratoire Eylau Unilabs, Paris; b Clinique de la Muette, Paris; c Clinique Pierre Che Paris ne tique, Centre Hospitalier Universitaire, Amiens; e INSERM, Universite Biologie de la Reproduction et Cytoge de Me decine, Assistance Publique–Ho ^ pitaux de Paris, Groupe Hospitalier Universitaire Ouest, Centre Descartes, Faculte cologie Obste trique II et Me decine de la Reproduction, Paris, Hospitalier Universitaire Cochin Port Royal, Service de Gyne France
Objective: To evaluate the influence of body mass index (BMI) on semen characteristics. Design: Cohort study. Setting: Single private andrology laboratory. Patient(s): All patients (n ¼ 10,665) consulting for a semen analysis from October 9, 2010, to October 8, 2011. When analyses were repeated on the same patient, only the first was included. Intervention(s): Recording of self-reported weight and height and of semen analysis. Main Outcome Measure(s): All parameters of standard semen analysis: pH, volume, sperm concentration per mL, total sperm count per ejaculate, motility (%) within 1 hour after ejaculation (overall and progressive), viability (%), and normal sperm morphology (%). Parametric and nonparametric statistical methods were applied, and results are given either with mean SD, or 10th, 50th, and 90th percentiles. Result(s): Semen volume decreased from 3.3 1.6 to 2.7 1.6 mL when BMI increased from normal (20–25 kg/m2) to extreme obesity (>40 kg/m2). The same was true for semen concentration (56.4 54.9 to 39.4 51.0 million/mL), total sperm count (171 170 to 92 95 million), and progressive motility (36.9 16.8% to 34.7 17.1%). The percentage of cases with azoospermia and cryptozoospermia increased from 1.9% to 9.1% and from 4.7% to 15.2%, respectively. The other semen characteristics were not affected. Multivariate models including age and abstinence duration confirmed these results. Conclusion(s): In this study, on a large patient sample size, increased BMI was associated with decreased semen quality, affecting volume, concentration, and motility. The percentage of Use your smartphone normal forms was not decreased. (Fertil SterilÒ 2014;102:1268–73. Ó2014 by American Society to scan this QR code for Reproductive Medicine.) and connect to the Key Words: BMI, semen volume, concentration, motility, morphology, obesity Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/bellocs-bmi-semen-parameters-morphology/
T
he impact of weight abnormalities on female fertility has been analyzed at length. High (>25 kg/m2) and low (<18 kg/m2) body mass index (BMI) have been
related to ovulation dysfunction, decreased fecundity, and lower pregnancy rates (1–3). In contrast, relatively few papers have addressed this question in men in the past, even
Received February 24, 2014; revised and accepted July 11, 2014; published online September 12, 2014. S.B. has nothing to disclose. M.C.-B. has nothing to disclose. E.A. has received fees for lectures from Lilly. V.I. has nothing to disclose. M.B. has nothing to disclose. A.D. has nothing to disclose. J.d.M. has received fees from Laboratoire Genevrier and Laboratoire Ferring. Reprint requests: Jacques de Mouzon, M.D., M.P.H., INSERM, 15 rue Guilleminot, Paris 75014, France (E-mail: [email protected]). Fertility and Sterility® Vol. 102, No. 5, November 2014 0015-0282/$36.00 Copyright ©2014 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2014.07.1212 1268
discussion forum for this article now.*
* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.
if literature is now growing. Recently, Relwani et al. (4) concluded that semen parameters are unrelated to BMI but vary with selective serotonin reuptake inhibitor (SSRI) use and previous urologic surgery. However, that study was conducted on a relatively small sample (530 men), and the relationship with a prediction of clinical pregnancy failed to reach significance (P¼ .06). In a systematic review with meta-analysis, MacDonald et al. (5) found only five out of 31 VOL. 102 NO. 5 / NOVEMBER 2014
Fertility and Sterility® studies to be suitable for analysis. They concluded that there was no evidence for a relationship between BMI and sperm concentration or total sperm count, but noted a negative relationship between increased BMI and testosterone, SHBG, and free testosterone levels. The main limitation was that data from most studies could not be aggregated for metaanalysis, and population-based studies with larger sample sizes and longitudinal studies were required. Recently, Shayeb et al. (6) published a study on 2,035 men, all partners of couples attending for infertility investigations in a single Fertility Clinic from 1990 to 2007. They found a statistically increased risk of low (<2 mL) semen volume (odds ratio [OR] 1.69, 95% confidence interval [CI] 1.20–2.38), and lower (<15%) normal sperm morphology (OR 1.50, 95% CI 1.06– 2.09) in obese (BMI R30 kg/m2) versus normal men, but no relationship with sperm concentration and motility. There was no significant relationship in men who were moderately overweight (BMI 25–30 kg/m2) or with low weight (BMI <18.5 kg/m2). Moreover, the large study duration (18 years) meant that only 38% of samples could be used owing to missing data on BMI. Finally, it was not mentioned if there were several sperm analyses performed for the same man during the study period and how they were handled. More recently, Colaci et al. (7) showed that fertilization rate was higher among obese men than among normal-weight men in conventional IVF cycles. No statistically significant associations were found between male BMI and the proportion of poor-quality embryos on day 3, slow embryo cleavage rate, or accelerated embryo cleavage rate. Male BMI was unrelated to positive b-hCG rate, clinical pregnancy rate, or live birth rate per embryo transfer. However, among couples undergoing intracytoplasmic sperm injection (ICSI), the ORs for live birth in couples with obese male partners was 84% lower than those where the men had normal BMI. The authors suggested a possible deleterious effect of male obesity on the chances of having a live birth among couples undergoing ICSI. Finally, MacDonald et al. (8), in 511 men, found a correlation only between BMI and normal sperm morphology. Barazani et al. (9) concluded from a literature review that sperm parameters would be adversely affected by diet intake, even if studies on the relationship between obesity and sperm quality was conflicting, and Palmer et al. (10) concluded from a literature review that results on this relationship were conflicting. Sermondade et al. (11), in a multinational metaanalysis on 13,077 men, concluded that overweight and obesity were associated with an increased prevalence of azoospermia or oligozoospermia, but with an important limitation due to the heterogeneity of the studied populations (both general population and infertile couples). They also emphasized that the possible impact of weight normalization on improvement of sperm parameters should be evaluated further. The aim of the present study was to evaluate the influence of BMI on semen characteristics in a large cohort of semen analyses performed in a single andrology laboratory during 1 year.
MATERIAL AND METHODS Patients and Variables Self-reported weight and height were routinely recorded for each man performing a semen analysis in our laboratory since VOL. 102 NO. 5 / NOVEMBER 2014
October 9, 2010. The present study investigated all analyses up to October 8, 2011. Men were referred to our laboratory in the course of a couple infertility evaluation of any origin. When more than one semen analysis had been performed for the same man, only the first one was selected for analysis, i.e., 10,665 out of 11,715. Age and abstinence period were also routinely recorded.
Semen Analysis Semen samples were collected in the laboratory by masturbation into a sterile container and men were instructed to respect a period of 2–7 days of abstinence beforehand. After liquefaction, semen analysis was carried out according to the latest World Health Organization (WHO) laboratory manual for the examination and processing of human sperm (12), except for morphology. A routine analysis included assessment of pH, semen volume (mL), sperm concentration (million/mL), total sperm count (million), percentage motility (overall and progressive), and percentage normal forms. Sperm morphology was assessed according to a modified David classification (13, 14) after Harris-Schorr staining. Concentration and total sperm count were quoted as ‘‘0’’ in cases of azoospermia. Motility and morphology were measured on all cases without azoospermia or cryptozoospermia. Patients with abstinence duration <2 days (1.9%) or >7 days (2.7%) or where abstinence was not reported (n ¼ 43) were excluded, leaving 10,197 patients for analysis.
Statistical Analysis All quantitative variables were reported with mean and standard deviation or median and 10th and 90th percentiles as appropriate. The study population was divided into six groups according to WHO BMI classification (15): underweight (<18.50 kg/m2), normal weight (18.50–24.99 kg/m2), overweight (25.00–29.99 kg/m2), moderate obesity (30.00– 34.99 kg/m2), severe obesity (35.00–39.99 kg/m2), and morbid obesity (R40.00 kg/m2). The relationships between BMI and semen characteristics were analyzed with the use of correlation coefficients for natural BMI and analysis of variance (ANOVA) for groups of BMI. Then, multivariate variance-covariance analysis (general linear model) was used to take into account age and abstinence duration. Sperm concentration and total sperm count were analyzed both with natural values and with their log transformations, as often proposed (16), as well as with parametric ANOVA and ANOVA on ranks, and with Kruskal-Wallis test. Age was divided into two categories, <40 and R40 years. Finally, a multilogistical model was used to analyze the risk of having the following: low volume (<1.5 mL), low concentration (<15 million/mL), low sperm count (<39 million), low progressive motility (<32%), according to WHO reference lower limits (5th percentiles) for semen characteristics (17) according to increasing BMI, and controlling for age and abstinence duration. In this model, ORs are given with their 95% CIs. Analysis was performed using SAS (Statistic Applied Software) 9.1.3 Service Pack 3, on the INSERM computer (Villejuif, France). 1269
ORIGINAL ARTICLE: ANDROLOGY Institutional Review Board approval was not necessary, because the study was anonymous and did not include any medical intervention.
RESULTS The mean age of patients was 37.1 6.1 years (range 17–72), the mean BMI 25.0 3.4 kg/m2 (range 15.9–59.7), and the mean abstinence duration 3.9 1.2 days (range 2–7).
Correlation between BMI and Semen Characteristics BMI was negatively and significantly correlated (Table 1) with semen volume, sperm concentration (log), total sperm count (log), vitality, and motility (overall and progressive). However, all coefficients were weak, even though very significant. There was no correlation with normal forms. Moreover, BMI was positively and significantly correlated with patient's age (r ¼ 0.10; P< .001). In terms of natural values, semen volume (Table 2) decreased from 3.3 1.6 mL to 2.7 1.6 mL (P< .001) when BMI increased from normal to extreme obesity. A similar decrease was observed for sperm concentration (from 56.4 54.9 million/mL to 39.4 51.0 million/mL; P< .001 with the log-transformed variable) and total sperm count (from 171 170 million to 92 95 million; P< .001 with the log-transformed variable). The decrease was less important for progressive motility (from 36.9 16.8% to 34.7 17.1%; P< .01), and there was no significant change for the other examined semen characteristics, including morphology. The percentage of azoospermia increased from 1.9% to 9.1% (P< .01) and the percentage of cryptozoospermia from 4.7% to 15.2% (P¼ .02), an overall combined increase from 6.6% to 24.3% (P< .001). Several semen characteristics were related to age and abstinence duration. Between the youngest (<25 years) and the oldest men (R55 years), semen volume decreased from 3.4 1.5 mL to 2.1 1.3 mL (P< .001), concentration increased from 51.7 56.6 million/mL to 57.8 72.3 million/mL (P< .05), total sperm count decreased from 170 222 million to 122 187 million, and progressive motility
TABLE 1 Correlation coefficients between BMI and patients' semen characteristics (n [ 10,665). Significance test r value P value
r value P value
Age
Abstinence duration
0.10 .001
0.01 .64
Concentration Volume log
pH
0.02 0.06 .08 .001 Motility
0.04 .01
Sperm count log
Vitality
Total
Progressive
Normal forms
0.05 .001
0.02 .04
0.03 .01
0.03 .01
0.01 .15
Note: Correlation coefficients are those given by Spearman's rank tests. BMI ¼ body mass index. Belloc. Obesity affects semen production. Fertil Steril 2014.
1270
decreased from 36.3 16.9% to 28.6 18.0% (P< .001). The percentage of normal forms varied without significant trend. Abstinence duration was not correlated with BMI (r ¼ 0.01; P¼ .64), but with semen characteristics. When duration increased from 1 to 7 days, volume increased from 2.8 1.4 mL to 3.5 1.8 mL (P< .001), concentration from 48.1 45.2 million/mL to 78.8 79.1 million/mL (P< .001), and total sperm count from 125 119 million to 250 242 million (P< .001). Abstinence duration only mildly influenced total motility (from 38.8 16.2% to 41.7 16.8%; P< .01), progressive motility (from 35.8 16.3% to 38.9 16.7%; P< .01) and did not drastically change the percentage of normal forms, even though significant (12.7 8.3% vs. 13.0 7.8%; P¼ .04).
Multivariate Modeling of the Role of BMI In multivariate models, taking into account age and abstinence duration (Table 3), volume and total sperm count were decreased with increased BMI, regardless of the chosen limit for BMI. Sperm concentration was also decreased for all limits, though nonsignificantly for extreme obesity. This was probably because of lower power due to small sample size (n ¼ 34). The decrease in progressive motility was less significant, except for obese men (BMI R30 kg/m2). In contrast, no significant differences were demonstrated for low BMI (<18.50 kg/m2), possibly owing to the low number of men (n ¼ 29). In the logistic multivariate models (Table 4), compared with normal weight (18.5–24.90 kg/m2), moderate obesity (30.0–34.9 kg/m2) was associated with significantly increased ORs for low volume (<1.5 mL: OR 1.42, 95% CI 1.11–1.82), low concentration (OR 1.40, 95% CI 1.16–1.68), low sperm count (OR 1.52; 95% CI 1.27–1.83), azoospermia or cryptozoospermia (OR 1.63, 95% CI 1.24–2.15), and low progressive motility (OR 1.29, 95% CI 1.08–1.53). Similar results were generally observed for severe and extreme obesity, but not always significantly, probably owing to lower sample sizes (n ¼ 97 and n ¼ 33, respectively). Simple overweight was not significantly associated to higher risks, even though the number of men provided high enough power (n ¼ 3,607). One model was also performed on teratozoospermia (<4% normal forms), which did not find any relationship with BMI, and only a higher risk for men older than 40 years (OR 1.16, 95% CI 1.03–1.32).
DISCUSSION This large single-clinic study was performed on >10,000 samples to investigate the relationship between male BMI and semen parameters. There was a clear association between sperm production (volume, concentration, and total sperm count) and obesity, whereas the impact on motility was questionable, and no statistically significant associations were found with other semen parameters, particularly sperm morphology. The strengths of this paper are that it relies on the largest sample size ever published by a single centre investigating the impact of BMI on semen parameters and that all the semen analyses were performed over a short time period. Three VOL. 102 NO. 5 / NOVEMBER 2014
Fertility and Sterility®
TABLE 2 Age, duration of abstinence, and sperm characteristics in relation to BMI. BMI (kg/m2)a Thin
Normal
Overweight
Obese
Severe obese
Extreme obese
<18.5
18.5–24.9
25.0–29.9
30.0–34.9
35.0–39.9
‡40.0
27 34.0 6.5 4.3 1.3 8.40 0.32 3.0 1.3 1.2–3.1–4.4 49.5 49.0 4–37–91 133 104 18–102–323 75.7 9.7
5799 36.4 6.4 3.9 1.2 8.41 0.32 3.3 1.6 1.5–3.1–5.4 56.4 54.9 4–42–122 171 170 9–125–391 74.1 12.6
3607 37.8 7.0 3.8 1.2 8.40 0.33 3.2 1.5 1.4–3.0–5.2 55.1 56.9 3–40–118 163 175 8–113–374 73.8 12.7
634 38.0 7.3 3.8 1.2 8.39 0.46 3.1 1.6 1.2–2.9–5.2 50.7 55.7 1–36–112 141 166 3–96–334 72.5 13.5
97 37.7 8.4 4.0 1.3 8.40 0.32 3.0 1.8 1.0–2.7–5.45 49.7 49.5 4–38–108 136 144 11–86–374 75.0 10.0
33 35.8 6.8 3.7 1.3 8.38 0.43 2.7 1.6 1.2–2.3–4.4 39.4 51.0 0.1–21–90 92 95 0.1–64–233 77.5 12.5
38.3 13.9 23–40–55 35.1 13.8 20–36–52 10.7 7.2 4–8–21
39.7 16.7 17–39–62 36.9 16.8 14–37–59 12.1 7.9 2–11–23
39.4 16.6 17–39–62 36.5 16.8 14–36–59 11.9 7.8 2–11–23
37.5 16.6 16–36–61 34.4 16.9 11–33–59 11.5 8.0 2–11–23
38.5 15.8 17–37–61 35.6 15.9 14–34–58 12.0 8.4 3–10–25
38.0 16.2 15–38–56 34.7 17.1 5–35–53 12.5 9.0 3–10–24
Characteristic N Age, y Abstinence, d pH Volume, mL 10–50–90 centiles Concentration, 106/mL 10–50–90 centiles Sperm count, millions 10–50–90 centiles Viability, % Motility, % Total 10–50–90 centiles Progressive 10–50–90 centiles Normal forms, % 10–50–90 centiles
P value .001 .05 .79 .001 .001 .001 .03 .05 .02 .42
Note: All values are expressed as mean SD and 10th, 50th, and 90th centiles for main characteristics. Analysis performed using ANOVA, with the log-transformed variable for concentration and sperm count. a BMI groups are partitioned according to WHO standards. BMI ¼ body mass index. Belloc. Obesity affects semen production. Fertil Steril 2014.
studies were published recently by Relwany et al. (4), Shayeb et al. (6), and Macdonald et al. (8) investigating 530, 2,035, and 511 men, respectively. Our study was also greater than the meta-analysis published by Cooper et al. (17) on 4,860 men. Only the meta-analysis by Sermondade et al. (11) was performed on a larger sample (13,000), but those authors acknowledged an important limitation due to the heterogeneity of the studied populations. Taking into account our population structure, the present work allowed us to get 80% power to detect a difference of 0.1 mL, 3 million sperm/mL, 10 million total sperm, 1% motility, and 0.5% normal forms between normal weight and all overweight with a two-sided 5% level of significance. For all obese patients, the study power was enough to detect differences of 0.17 mL, 5 million/mL, 19 million, 2.8%, and 1%, respectively, for BMI >30 kg/m2,
TABLE 3 The statistical relationship between semen characteristics and each abnormal BMI class against normal BMI in multivariate models (general linear model), taking in account age and abstinence duration. BMI (kg/m2) Characteristic
>25
>30
>35
>40
<18.5
Volume Concentration Total sperm count Total motility Progressive motility Normal forms
0.001 0.05 0.001 0.18 0.10 0.26
0.001 0.001 0.001 0.01 0.01 0.10
0.01 0.03 0.001 0.33 0.30 0.55
0.02 0.12 0.01 0.91 0.78 0.77
0.19 0.42 0.20 0.28 0.28 0.25
Note: Values as expressed as P values. In each column, semen values observed in the specified BMI group are compared to normal BMI (18.5–24.9). BMI ¼ body mass index. Belloc. Obesity affects semen production. Fertil Steril 2014.
VOL. 102 NO. 5 / NOVEMBER 2014
of 0.4 mL, 14 million/mL, 42 million, 4.1%, and 2% for BMI >35 kg/m2, and 0.8 mL, 26 million/mL, 85 million, 8%, and 4% for BMI >40 kg/m2. Finally, for underweight men, the detectable differences were 0.8 mL, 30 million/mL, 90 million, 9%, and 4%, respectively. Thus, the power was sufficient to analyze overweight and obesity, and moderate for severe and extreme obesity in case of relatively small impact. Another advantage of this large sample was to allow for multivariate analysis, taking into account major confounders such as male age and abstinence duration. However, it was not possible to analyze other potential confounders, as sociodemographic characteristics and smoking, because this information was not collected in the database. In addition, the study of such a large sample size collected in a short period (1 year) is a greater guarantee of homogeneity in technical procedures. This is in contrast to the Shayeb et al. (6) paper, where data were collected over 18 years. The use of a modified David classification (13, 14) in our laboratory for morphology (18) is based on experience and constitutes a standard method in France (19), even if it does not really allow for international comparisons on morphology. However, all other sperm parameters were analyzed according to WHO manual (12) and can be compared. The results of our study are in accordance with some already published data showing that semen in obese men is of poorer quality than in men of normal weight (20, 21). However, there are some differences. Compared with Shayeb et al. (6), we found a similar relationship with semen volume but also an impact of BMI on concentration and sperm count and a trending impact on progressive motility. On the other hand, there was no correlation between BMI and sperm morphology. 1271
ORIGINAL ARTICLE: ANDROLOGY
TABLE 4 Impact of BMI on semen anomalies. Parameter N BMI, kg/m2 <18.5 18.5–24.99 25.0–29.99 30.0–34.99 35.0–39.99 R40.0 Age, y R40 vs.<40 Abstinence duration
Ejaculate volume, <1.5 mL
Concentration, <15 millions/mL
Total sperm count, <39 millions
Azoo- or crypto-zoospermia
Progressive motility, <32%
10,195
10,195
10,195
10,195
9,755
2.42 (0.90–6.48) 1.00 1.08 (0.94–1.24) 1.42 (1.11–1.82) 2.39 (1.44–3.96) 2.07 (0.84–5.07)
0.45 (0.14–1.51) 1.00 1.04 (0.93–1.14) 1.40 (1.16–1.68) 0.95 (0.59–1.55) 1.93 (0.95–3.94)
0.67 (0.23–1.93) 1.00 1.06 (0.96–1.17) 1.52 (1.27–1.83) 1.17 (0.74–1.87) 2.01 (0.99–4.11)
0.56 (0.08–4.14) 1.00 1.05 (0.89–1.24) 1.63 (1.24–2.15) 0.61 (0.22–1.66) 4.51 (2.02–10.08)
1.17 (0.54–2.54) 1.00 1.04 (0.95–1.13) 1.29 (1.08–1.53) 1.15 (0.76–1.74) 1.03 (0.47–2.27)
2.01 (1.76–2.29) 0.88 (0.83–0.93)
1.16 (1.05–1.28) 0.91 (0.87–0.95)
1.32 (1.19–1.46) 0.90 (0.86–0.93)
1.18 (1.00–1.39) 0.97 (0.91–1.04)
1.32 (1.20–1.44) 0.96 (0.93–0.99)
Note: Values are expressed as odds ratio (95% confidence interval). One multivariate model per semen parameter. BMI ¼ body mass index. Belloc. Obesity affects semen production. Fertil Steril 2014.
Our results conflict with those published by Relwani et al. (4), who did not find any impact of BMI on semen characteristics. They also contradict the recent meta-analysis (5), but it can be noted that those authors found a negative relationship for T, SHBG, and free T (21, 22) with increased BMI, which may, at least in part, explain our findings. Similar findings were reported by Al-Ali et al. (23), with no influence on all assessed sperm quality parameters, but on LH (P¼ .001), T (P< .001), and PRL (P¼ .044). In a cohort of Danish men, serum T, SHBG, and inhibin B all decreased with increasing BMI, whereas free androgen index and E2 increased with increasing BMI. Serum FSH was higher among slim men (24). One could speculate that the effect of obesity on clinical outcomes may be mediated through semen quality. However, the Colaci et al. analysis (7) indicated that the relationship between male obesity and live birth rates was independent from sperm parameters. No data suggest any plausible biologic explanation for the higher fertilization rate and higher live birth observed with assisted reproductive technologies among couples with an obese man in that study. The limited sample size, and the potential effect of other confounders, such as female age, have to be examined in future studies. Interestingly, Ramlau-Hansen et al. in 2007 (25) published a study in which the risk of subfecundity (time to achieve pregnancy >12 months) was estimated in 47,835 pregnancies according to the body weight of both members of the couple. The OR for subfecundity increased between underweight and obese men and women in a dose-response manner. Obesity has been related also to a more than twofold risk of sperm DNA damage (OR 2.5, 95% CI 1.2–5.1) on 330 men in infertile couples (26). Another study (27) found a small increase of oxidative stress with increasing BMI, even if the magnitude of this increase was of minor clinical significance, without an associated decline in sperm DNA integrity or sperm motility with increasing reactive oxygen species production. Our results also contradicted those of MacDonald et al. (8), who did not find an effect on concentration or motility of overweight, but a light impact on morphology. However, their study relied on a relatively small number, with a power only able to detect at least a doubling in abnormally low 1272
sperm concentration and total sperm count (OR R2.0) for overweight or obese men compared with men with a normal BMI. One limitation of our study is that it was conducted among men seeking fertility treatment. Moreover, to improve our knowledge about the effect of weight or obese status on reproduction, researchers should always register for both men and women not only their BMI but also factors associated with weight excess that could also affect fertility, such as lifestyle habits, fat distribution, and associated pathologies. For example, Hammiche et al. (16) found that not only BMI, but also waist circumference R102 cm, a measure for central adiposity, was inversely associated with sperm concentration and total motile sperm count (P< .001 and P< .02, respectively). This was confirmed by a recent paper on 501 men (28). It is also interesting to note that H akonsen et al. (22), on a relatively low sample size, concluded that weight loss could improve semen quality. How increased BMI can alter semen quality has been the subject of a number of studies in nonhuman animal models. A recent study (29) investigated the effect of diet-induced paternal obesity, in the absence of diabetes, on the metabolic health of two resultant generations of mice and concluded that it can modulate sperm microRNA content and germ cell methylation status. Similarly, a study on rats (30) stated, ‘‘observations demonstrate [that] hydrocarbons can promote epigenetic transgenerational inheritance of disease and sperm epimutations, potential biomarkers for ancestral exposure.’’ Induced obesity may also lead to decreased sperm reserves and acceleration of transit time in the epididymis of adult male rats (30). Conversely, diet and exercise may reverse perturbed sperm function (31), and Palmer et al. (32) proposed a role for mammalian SIRT6 protein in spermatogenesis, which is known to be compromised by male obesity. Although it is difficult to generalize our results to the overall population, our results suggest that increased BMI has a deleterious effect on semen concentration but not necessarily on reproductive outcome. Other studies are needed to analyze the benefit of weight reduction among obese men. VOL. 102 NO. 5 / NOVEMBER 2014
Fertility and Sterility® Acknowledgments: The authors thank Denny Sakkas for assistance in reviewing the manuscript.
REFERENCES 1.
2. 3.
4.
5.
6.
7.
8.
9. 10. 11.
12. 13.
14.
15. 16.
17.
Balen AH, Anderson RA, Policy and Practice Committee of the BFS. Impact of obesity on female reproductive health: British Fertility Society, policy and practice guidelines. Hum Fertil 2007;10:195–206. Pasquali R, Patton L, Gambineri A. Obesity and infertility. Curr Opin Endocrinol Diabetes Obes 2007;14:482–7. Maheshwari A, Stofberg L, Bhattacharya S. Effect of overweight and obesity on assisted reproductive technology—a systematic review. Hum Reprod Update 2007;13:433–44. Relwani R, Berger D, Santoro N, Hickmon C, Nihsen M, Zapantis A, et al. Semen parameters are unrelated to BMI but vary with SSRI use and prior urological surgery. Reprod Sci 2011;18:391–7. MacDonald AA, Herbison GP, Showell M, Farquhar CM. The impact of body mass index on semen parameters and reproductive hormones in human males: a systematic review with meta-analysis. Hum Reprod Update 2010; 16:293–311. Shayeb AG, Harrild K, Mathers E, Bhattacharya S. An exploration of the association between male body mass index and semen quality. Reprod Biomed Online 2011;23:717–23. Colaci DS, Afeiche M, Gaskins AJ, Wright DL, Toth TL, Tanrikut C, et al. Men's body mass index in relation to embryo quality and clinical outcomes in couples undergoing in vitro fertilization. Fertil Steril 2012;98:1193–9. Macdonald AA, Stewart AW, Farquhar CM. Body mass index in relation to semen quality and reproductive hormones in New Zealand men: a crosssectional study in fertility clinics. Hum Reprod 2013;28:3178–87. Barazani Y, Katz BF, Nagler HM, Stember DS. Lifestyle, environment, and male reproductive health. Urol Clin North Am 2014;41:55–66. Palmer NO, Bakos HW, Fullston T, Lane M. Impact of obesity on male fertility, sperm function and molecular composition. Spermatogenesis 2012;2:1–11. Sermondade N, Faure C, Fezeu L, Shaye AG, Bonde JB, Jensen TK, et al. BMI in relation to sperm count: an updated systematic review and collaborative meta-analysis. Hum Reprod Update 2013;19:221–31. WHO laboratory manual for the examination and processing of human sperm. 5th ed. Geneva: World Health Organization; 2010. Auger J, Eustache F. Standardization of the morphological assessment of human spermatozoa according to modified David's classification. Andrologie 2000;10:353–87. Blanchard M, Haguenoer K, Apert A, Poret H, Barthelemy C, Royere D, et al. Sperm morphology assessment using David's classification: time to switch to strict criteria? Prospective comparative analysis in a selected IVF population. Int J Androl 2011;2:145–52. World Health Organisation. BMI classifications. Available from: www.who. int/bmi/index.jsp?introPage¼intro_3.html; 2006. Hammiche F, Laven JS, Twigt JM, Boellaard WP, Steegers EA, SteegersTheunissen RP. Body mass index and central adiposity are associated with sperm quality in men of subfertile couples. Hum Reprod 2012;27:2365–72. Cooper TG, Noonan E, von Eckardstein S, Auger J, Baker HW, Behre HM, et al. World Health Organization reference values for human semen characteristics. Hum Reprod Update 2010;16:231–45.
VOL. 102 NO. 5 / NOVEMBER 2014
18.
19.
20.
21.
22.
23.
24.
25.
26.
27. 28.
29.
30.
31.
32.
Prisant N, Cohen-Bacrie P, Amar E, Belaisch-Allart J, Cohen-Bacrie M, Olivennes F, et al. Teratozoospermia, myth or reality? A 10-years retrospective study on 101404 consecutive sperm samples. Gynecol Obstet Fertil 2011;39:136–40. Rolland M, le Moal J, Wagner V, Royere D, de Mouzon J. Decline in semen concentration and morphology in a sample of 26609 men close to general population between 1989 and 2005 in France. Hum Reprod 2013;28:462–70. Chavarro JE, Toth TL, Wright DL, Meeker JD, Hauser R. Body mass index in relation to semen quality, sperm DNA integrity, and serum reproductive hormone levels among men attending an infertility clinic. Fertil Steril 2010;93:2222–31. Zumoff B, Strain GW, Miller LK, Rosner W, Senie R, Seres DS, et al. Plasma free and nonsex-hormone-binding-globulin-bound testosterone are decreased in obese men in proportion to their degree of obesity. J Clin Endocrinol Metab 1990;71:929–31. H akonsen LB, Thulstrup AM, Aggerholm AS, Olsen J, Bonde JP, Andersen CY, et al. Does weight loss improve semen quality and reproductive hormones? Results from a cohort of severely obese men. Reprod Health 2011;8:24. M Al-Ali B, Gutschi T, Pummer K, Zigeuner R, Brookman-May S, Wieland WF, et al. Body mass index has no impact on sperm quality but on reproductive hormones levels. Andrologia 2014;46:106–11. Jensen TK, Andersson AM, Jørgensen N, Andersen AG, Carlsen E, Petersen JH, et al. Body mass index in relation to semen quality and reproductive hormones among 1,558 Danish men. Fertil Steril 2004;82:863–70. Ramlau-Hansen CH, Thulstrup AM, Nohr EA, Bonde JP, Sorensen TI, Olsen J. Subfecundity in overweight and obese couples. Hum Reprod 2007;22: 1634–7. Dupont C, Faure C, Sermondade N, Boubaya M, Eustache F, Clement P, et al. Obesity leads to higher risk of sperm DNA damage in infertile patients. Asian J Androl 2013;15:622–5. Tunc O, Bakos HW, Tremellen K. Impact of body mass index on seminal oxidative stress. Andrologia 2011;43:121–8. Eisenberg ML, Kim S, Chen Z, Sundaram R, Schisterman EF, Buck Louis GM. The relationship between male BMI and waist circumference on semen quality: data from the LIFE study. Hum Reprod 2014;29:193–200. Fullston T, Ohlsson Teague EM, Palmer NO, DeBlasio MJ, Mitchell M, Corbett M, et al. Paternal obesity initiates metabolic disturbances in two generations of mice with incomplete penetrance to the F2 generation and alters the transcriptional profile of testis and sperm microRNA content. FASEB J 2013;27:4226–43. Tracey R, Manikkam M, Guerrero-Bosagna C, Skinner MK. Hydrocarbons (jet fuel JP-8) induce epigenetic transgenerational inheritance of obesity, reproductive disease and sperm epimutations. Reprod Toxicol 2013;36: 104–16. Palmer NO, Bakos HW, Owens JA, Setchell BP, Lane M. Diet and exercise in an obese mouse fed a high-fat diet improve metabolic health and reverse perturbed sperm function. Am J Physiol Endocrinol Metab 2012;302: E768–80. Palmer NO, Fullston T, Mitchell M, Setchell BP, Lane M. SIRT6 in mouse spermatogenesis is modulated by diet-induced obesity. Reprod Fertil Dev 2011; 23:929–39.
1273
Objective: To evaluate the influence of body mass index (BMI) on semen characteristics. Design: Cohort study. Setting: Single private andrology laboratory. Patient(s): All patients (n ¼ 10,665) consulting for a semen analysis from October 9, 2010, to October 8, 2011. When analyses were repeated on the same patient, only the first was included. Intervention(s): Recording of self-reported weight and height and of semen analysis. Main Outcome Measure(s): All parameters of standard semen analysis: pH, volume, sperm concentration per mL, total sperm count per ejaculate, motility (%) within 1 hour after ejaculation (overall and progressive), viability (%), and normal sperm morphology (%). Parametric and nonparametric statistical methods were applied, and results are given either with mean SD, or 10th, 50th, and 90th percentiles. Result(s): Semen volume decreased from 3.3 1.6 to 2.7 1.6 mL when BMI increased from normal (20–25 kg/m2) to extreme obesity (>40 kg/m2). The same was true for semen concentration (56.4 54.9 to 39.4 51.0 million/mL), total sperm count (171 170 to 92 95 million), and progressive motility (36.9 16.8% to 34.7 17.1%). The percentage of cases with azoospermia and cryptozoospermia increased from 1.9% to 9.1% and from 4.7% to 15.2%, respectively. The other semen characteristics were not affected. Multivariate models including age and abstinence duration confirmed these results. Conclusion(s): In this study, on a large patient sample size, increased BMI was associated with decreased semen quality, affecting volume, concentration, and motility. The percentage of Use your smartphone normal forms was not decreased. (Fertil SterilÒ 2014;102:1268–73. Ó2014 by American Society to scan this QR code for Reproductive Medicine.) and connect to the Key Words: BMI, semen volume, concentration, motility, morphology, obesity Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/bellocs-bmi-semen-parameters-morphology/
T
he impact of weight abnormalities on female fertility has been analyzed at length. High (>25 kg/m2) and low (<18 kg/m2) body mass index (BMI) have been
related to ovulation dysfunction, decreased fecundity, and lower pregnancy rates (1–3). In contrast, relatively few papers have addressed this question in men in the past, even
Received February 24, 2014; revised and accepted July 11, 2014; published online September 12, 2014. S.B. has nothing to disclose. M.C.-B. has nothing to disclose. E.A. has received fees for lectures from Lilly. V.I. has nothing to disclose. M.B. has nothing to disclose. A.D. has nothing to disclose. J.d.M. has received fees from Laboratoire Genevrier and Laboratoire Ferring. Reprint requests: Jacques de Mouzon, M.D., M.P.H., INSERM, 15 rue Guilleminot, Paris 75014, France (E-mail: [email protected]). Fertility and Sterility® Vol. 102, No. 5, November 2014 0015-0282/$36.00 Copyright ©2014 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2014.07.1212 1268
discussion forum for this article now.*
* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.
if literature is now growing. Recently, Relwani et al. (4) concluded that semen parameters are unrelated to BMI but vary with selective serotonin reuptake inhibitor (SSRI) use and previous urologic surgery. However, that study was conducted on a relatively small sample (530 men), and the relationship with a prediction of clinical pregnancy failed to reach significance (P¼ .06). In a systematic review with meta-analysis, MacDonald et al. (5) found only five out of 31 VOL. 102 NO. 5 / NOVEMBER 2014
Fertility and Sterility® studies to be suitable for analysis. They concluded that there was no evidence for a relationship between BMI and sperm concentration or total sperm count, but noted a negative relationship between increased BMI and testosterone, SHBG, and free testosterone levels. The main limitation was that data from most studies could not be aggregated for metaanalysis, and population-based studies with larger sample sizes and longitudinal studies were required. Recently, Shayeb et al. (6) published a study on 2,035 men, all partners of couples attending for infertility investigations in a single Fertility Clinic from 1990 to 2007. They found a statistically increased risk of low (<2 mL) semen volume (odds ratio [OR] 1.69, 95% confidence interval [CI] 1.20–2.38), and lower (<15%) normal sperm morphology (OR 1.50, 95% CI 1.06– 2.09) in obese (BMI R30 kg/m2) versus normal men, but no relationship with sperm concentration and motility. There was no significant relationship in men who were moderately overweight (BMI 25–30 kg/m2) or with low weight (BMI <18.5 kg/m2). Moreover, the large study duration (18 years) meant that only 38% of samples could be used owing to missing data on BMI. Finally, it was not mentioned if there were several sperm analyses performed for the same man during the study period and how they were handled. More recently, Colaci et al. (7) showed that fertilization rate was higher among obese men than among normal-weight men in conventional IVF cycles. No statistically significant associations were found between male BMI and the proportion of poor-quality embryos on day 3, slow embryo cleavage rate, or accelerated embryo cleavage rate. Male BMI was unrelated to positive b-hCG rate, clinical pregnancy rate, or live birth rate per embryo transfer. However, among couples undergoing intracytoplasmic sperm injection (ICSI), the ORs for live birth in couples with obese male partners was 84% lower than those where the men had normal BMI. The authors suggested a possible deleterious effect of male obesity on the chances of having a live birth among couples undergoing ICSI. Finally, MacDonald et al. (8), in 511 men, found a correlation only between BMI and normal sperm morphology. Barazani et al. (9) concluded from a literature review that sperm parameters would be adversely affected by diet intake, even if studies on the relationship between obesity and sperm quality was conflicting, and Palmer et al. (10) concluded from a literature review that results on this relationship were conflicting. Sermondade et al. (11), in a multinational metaanalysis on 13,077 men, concluded that overweight and obesity were associated with an increased prevalence of azoospermia or oligozoospermia, but with an important limitation due to the heterogeneity of the studied populations (both general population and infertile couples). They also emphasized that the possible impact of weight normalization on improvement of sperm parameters should be evaluated further. The aim of the present study was to evaluate the influence of BMI on semen characteristics in a large cohort of semen analyses performed in a single andrology laboratory during 1 year.
MATERIAL AND METHODS Patients and Variables Self-reported weight and height were routinely recorded for each man performing a semen analysis in our laboratory since VOL. 102 NO. 5 / NOVEMBER 2014
October 9, 2010. The present study investigated all analyses up to October 8, 2011. Men were referred to our laboratory in the course of a couple infertility evaluation of any origin. When more than one semen analysis had been performed for the same man, only the first one was selected for analysis, i.e., 10,665 out of 11,715. Age and abstinence period were also routinely recorded.
Semen Analysis Semen samples were collected in the laboratory by masturbation into a sterile container and men were instructed to respect a period of 2–7 days of abstinence beforehand. After liquefaction, semen analysis was carried out according to the latest World Health Organization (WHO) laboratory manual for the examination and processing of human sperm (12), except for morphology. A routine analysis included assessment of pH, semen volume (mL), sperm concentration (million/mL), total sperm count (million), percentage motility (overall and progressive), and percentage normal forms. Sperm morphology was assessed according to a modified David classification (13, 14) after Harris-Schorr staining. Concentration and total sperm count were quoted as ‘‘0’’ in cases of azoospermia. Motility and morphology were measured on all cases without azoospermia or cryptozoospermia. Patients with abstinence duration <2 days (1.9%) or >7 days (2.7%) or where abstinence was not reported (n ¼ 43) were excluded, leaving 10,197 patients for analysis.
Statistical Analysis All quantitative variables were reported with mean and standard deviation or median and 10th and 90th percentiles as appropriate. The study population was divided into six groups according to WHO BMI classification (15): underweight (<18.50 kg/m2), normal weight (18.50–24.99 kg/m2), overweight (25.00–29.99 kg/m2), moderate obesity (30.00– 34.99 kg/m2), severe obesity (35.00–39.99 kg/m2), and morbid obesity (R40.00 kg/m2). The relationships between BMI and semen characteristics were analyzed with the use of correlation coefficients for natural BMI and analysis of variance (ANOVA) for groups of BMI. Then, multivariate variance-covariance analysis (general linear model) was used to take into account age and abstinence duration. Sperm concentration and total sperm count were analyzed both with natural values and with their log transformations, as often proposed (16), as well as with parametric ANOVA and ANOVA on ranks, and with Kruskal-Wallis test. Age was divided into two categories, <40 and R40 years. Finally, a multilogistical model was used to analyze the risk of having the following: low volume (<1.5 mL), low concentration (<15 million/mL), low sperm count (<39 million), low progressive motility (<32%), according to WHO reference lower limits (5th percentiles) for semen characteristics (17) according to increasing BMI, and controlling for age and abstinence duration. In this model, ORs are given with their 95% CIs. Analysis was performed using SAS (Statistic Applied Software) 9.1.3 Service Pack 3, on the INSERM computer (Villejuif, France). 1269
ORIGINAL ARTICLE: ANDROLOGY Institutional Review Board approval was not necessary, because the study was anonymous and did not include any medical intervention.
RESULTS The mean age of patients was 37.1 6.1 years (range 17–72), the mean BMI 25.0 3.4 kg/m2 (range 15.9–59.7), and the mean abstinence duration 3.9 1.2 days (range 2–7).
Correlation between BMI and Semen Characteristics BMI was negatively and significantly correlated (Table 1) with semen volume, sperm concentration (log), total sperm count (log), vitality, and motility (overall and progressive). However, all coefficients were weak, even though very significant. There was no correlation with normal forms. Moreover, BMI was positively and significantly correlated with patient's age (r ¼ 0.10; P< .001). In terms of natural values, semen volume (Table 2) decreased from 3.3 1.6 mL to 2.7 1.6 mL (P< .001) when BMI increased from normal to extreme obesity. A similar decrease was observed for sperm concentration (from 56.4 54.9 million/mL to 39.4 51.0 million/mL; P< .001 with the log-transformed variable) and total sperm count (from 171 170 million to 92 95 million; P< .001 with the log-transformed variable). The decrease was less important for progressive motility (from 36.9 16.8% to 34.7 17.1%; P< .01), and there was no significant change for the other examined semen characteristics, including morphology. The percentage of azoospermia increased from 1.9% to 9.1% (P< .01) and the percentage of cryptozoospermia from 4.7% to 15.2% (P¼ .02), an overall combined increase from 6.6% to 24.3% (P< .001). Several semen characteristics were related to age and abstinence duration. Between the youngest (<25 years) and the oldest men (R55 years), semen volume decreased from 3.4 1.5 mL to 2.1 1.3 mL (P< .001), concentration increased from 51.7 56.6 million/mL to 57.8 72.3 million/mL (P< .05), total sperm count decreased from 170 222 million to 122 187 million, and progressive motility
TABLE 1 Correlation coefficients between BMI and patients' semen characteristics (n [ 10,665). Significance test r value P value
r value P value
Age
Abstinence duration
0.10 .001
0.01 .64
Concentration Volume log
pH
0.02 0.06 .08 .001 Motility
0.04 .01
Sperm count log
Vitality
Total
Progressive
Normal forms
0.05 .001
0.02 .04
0.03 .01
0.03 .01
0.01 .15
Note: Correlation coefficients are those given by Spearman's rank tests. BMI ¼ body mass index. Belloc. Obesity affects semen production. Fertil Steril 2014.
1270
decreased from 36.3 16.9% to 28.6 18.0% (P< .001). The percentage of normal forms varied without significant trend. Abstinence duration was not correlated with BMI (r ¼ 0.01; P¼ .64), but with semen characteristics. When duration increased from 1 to 7 days, volume increased from 2.8 1.4 mL to 3.5 1.8 mL (P< .001), concentration from 48.1 45.2 million/mL to 78.8 79.1 million/mL (P< .001), and total sperm count from 125 119 million to 250 242 million (P< .001). Abstinence duration only mildly influenced total motility (from 38.8 16.2% to 41.7 16.8%; P< .01), progressive motility (from 35.8 16.3% to 38.9 16.7%; P< .01) and did not drastically change the percentage of normal forms, even though significant (12.7 8.3% vs. 13.0 7.8%; P¼ .04).
Multivariate Modeling of the Role of BMI In multivariate models, taking into account age and abstinence duration (Table 3), volume and total sperm count were decreased with increased BMI, regardless of the chosen limit for BMI. Sperm concentration was also decreased for all limits, though nonsignificantly for extreme obesity. This was probably because of lower power due to small sample size (n ¼ 34). The decrease in progressive motility was less significant, except for obese men (BMI R30 kg/m2). In contrast, no significant differences were demonstrated for low BMI (<18.50 kg/m2), possibly owing to the low number of men (n ¼ 29). In the logistic multivariate models (Table 4), compared with normal weight (18.5–24.90 kg/m2), moderate obesity (30.0–34.9 kg/m2) was associated with significantly increased ORs for low volume (<1.5 mL: OR 1.42, 95% CI 1.11–1.82), low concentration (OR 1.40, 95% CI 1.16–1.68), low sperm count (OR 1.52; 95% CI 1.27–1.83), azoospermia or cryptozoospermia (OR 1.63, 95% CI 1.24–2.15), and low progressive motility (OR 1.29, 95% CI 1.08–1.53). Similar results were generally observed for severe and extreme obesity, but not always significantly, probably owing to lower sample sizes (n ¼ 97 and n ¼ 33, respectively). Simple overweight was not significantly associated to higher risks, even though the number of men provided high enough power (n ¼ 3,607). One model was also performed on teratozoospermia (<4% normal forms), which did not find any relationship with BMI, and only a higher risk for men older than 40 years (OR 1.16, 95% CI 1.03–1.32).
DISCUSSION This large single-clinic study was performed on >10,000 samples to investigate the relationship between male BMI and semen parameters. There was a clear association between sperm production (volume, concentration, and total sperm count) and obesity, whereas the impact on motility was questionable, and no statistically significant associations were found with other semen parameters, particularly sperm morphology. The strengths of this paper are that it relies on the largest sample size ever published by a single centre investigating the impact of BMI on semen parameters and that all the semen analyses were performed over a short time period. Three VOL. 102 NO. 5 / NOVEMBER 2014
Fertility and Sterility®
TABLE 2 Age, duration of abstinence, and sperm characteristics in relation to BMI. BMI (kg/m2)a Thin
Normal
Overweight
Obese
Severe obese
Extreme obese
<18.5
18.5–24.9
25.0–29.9
30.0–34.9
35.0–39.9
‡40.0
27 34.0 6.5 4.3 1.3 8.40 0.32 3.0 1.3 1.2–3.1–4.4 49.5 49.0 4–37–91 133 104 18–102–323 75.7 9.7
5799 36.4 6.4 3.9 1.2 8.41 0.32 3.3 1.6 1.5–3.1–5.4 56.4 54.9 4–42–122 171 170 9–125–391 74.1 12.6
3607 37.8 7.0 3.8 1.2 8.40 0.33 3.2 1.5 1.4–3.0–5.2 55.1 56.9 3–40–118 163 175 8–113–374 73.8 12.7
634 38.0 7.3 3.8 1.2 8.39 0.46 3.1 1.6 1.2–2.9–5.2 50.7 55.7 1–36–112 141 166 3–96–334 72.5 13.5
97 37.7 8.4 4.0 1.3 8.40 0.32 3.0 1.8 1.0–2.7–5.45 49.7 49.5 4–38–108 136 144 11–86–374 75.0 10.0
33 35.8 6.8 3.7 1.3 8.38 0.43 2.7 1.6 1.2–2.3–4.4 39.4 51.0 0.1–21–90 92 95 0.1–64–233 77.5 12.5
38.3 13.9 23–40–55 35.1 13.8 20–36–52 10.7 7.2 4–8–21
39.7 16.7 17–39–62 36.9 16.8 14–37–59 12.1 7.9 2–11–23
39.4 16.6 17–39–62 36.5 16.8 14–36–59 11.9 7.8 2–11–23
37.5 16.6 16–36–61 34.4 16.9 11–33–59 11.5 8.0 2–11–23
38.5 15.8 17–37–61 35.6 15.9 14–34–58 12.0 8.4 3–10–25
38.0 16.2 15–38–56 34.7 17.1 5–35–53 12.5 9.0 3–10–24
Characteristic N Age, y Abstinence, d pH Volume, mL 10–50–90 centiles Concentration, 106/mL 10–50–90 centiles Sperm count, millions 10–50–90 centiles Viability, % Motility, % Total 10–50–90 centiles Progressive 10–50–90 centiles Normal forms, % 10–50–90 centiles
P value .001 .05 .79 .001 .001 .001 .03 .05 .02 .42
Note: All values are expressed as mean SD and 10th, 50th, and 90th centiles for main characteristics. Analysis performed using ANOVA, with the log-transformed variable for concentration and sperm count. a BMI groups are partitioned according to WHO standards. BMI ¼ body mass index. Belloc. Obesity affects semen production. Fertil Steril 2014.
studies were published recently by Relwany et al. (4), Shayeb et al. (6), and Macdonald et al. (8) investigating 530, 2,035, and 511 men, respectively. Our study was also greater than the meta-analysis published by Cooper et al. (17) on 4,860 men. Only the meta-analysis by Sermondade et al. (11) was performed on a larger sample (13,000), but those authors acknowledged an important limitation due to the heterogeneity of the studied populations. Taking into account our population structure, the present work allowed us to get 80% power to detect a difference of 0.1 mL, 3 million sperm/mL, 10 million total sperm, 1% motility, and 0.5% normal forms between normal weight and all overweight with a two-sided 5% level of significance. For all obese patients, the study power was enough to detect differences of 0.17 mL, 5 million/mL, 19 million, 2.8%, and 1%, respectively, for BMI >30 kg/m2,
TABLE 3 The statistical relationship between semen characteristics and each abnormal BMI class against normal BMI in multivariate models (general linear model), taking in account age and abstinence duration. BMI (kg/m2) Characteristic
>25
>30
>35
>40
<18.5
Volume Concentration Total sperm count Total motility Progressive motility Normal forms
0.001 0.05 0.001 0.18 0.10 0.26
0.001 0.001 0.001 0.01 0.01 0.10
0.01 0.03 0.001 0.33 0.30 0.55
0.02 0.12 0.01 0.91 0.78 0.77
0.19 0.42 0.20 0.28 0.28 0.25
Note: Values as expressed as P values. In each column, semen values observed in the specified BMI group are compared to normal BMI (18.5–24.9). BMI ¼ body mass index. Belloc. Obesity affects semen production. Fertil Steril 2014.
VOL. 102 NO. 5 / NOVEMBER 2014
of 0.4 mL, 14 million/mL, 42 million, 4.1%, and 2% for BMI >35 kg/m2, and 0.8 mL, 26 million/mL, 85 million, 8%, and 4% for BMI >40 kg/m2. Finally, for underweight men, the detectable differences were 0.8 mL, 30 million/mL, 90 million, 9%, and 4%, respectively. Thus, the power was sufficient to analyze overweight and obesity, and moderate for severe and extreme obesity in case of relatively small impact. Another advantage of this large sample was to allow for multivariate analysis, taking into account major confounders such as male age and abstinence duration. However, it was not possible to analyze other potential confounders, as sociodemographic characteristics and smoking, because this information was not collected in the database. In addition, the study of such a large sample size collected in a short period (1 year) is a greater guarantee of homogeneity in technical procedures. This is in contrast to the Shayeb et al. (6) paper, where data were collected over 18 years. The use of a modified David classification (13, 14) in our laboratory for morphology (18) is based on experience and constitutes a standard method in France (19), even if it does not really allow for international comparisons on morphology. However, all other sperm parameters were analyzed according to WHO manual (12) and can be compared. The results of our study are in accordance with some already published data showing that semen in obese men is of poorer quality than in men of normal weight (20, 21). However, there are some differences. Compared with Shayeb et al. (6), we found a similar relationship with semen volume but also an impact of BMI on concentration and sperm count and a trending impact on progressive motility. On the other hand, there was no correlation between BMI and sperm morphology. 1271
ORIGINAL ARTICLE: ANDROLOGY
TABLE 4 Impact of BMI on semen anomalies. Parameter N BMI, kg/m2 <18.5 18.5–24.99 25.0–29.99 30.0–34.99 35.0–39.99 R40.0 Age, y R40 vs.<40 Abstinence duration
Ejaculate volume, <1.5 mL
Concentration, <15 millions/mL
Total sperm count, <39 millions
Azoo- or crypto-zoospermia
Progressive motility, <32%
10,195
10,195
10,195
10,195
9,755
2.42 (0.90–6.48) 1.00 1.08 (0.94–1.24) 1.42 (1.11–1.82) 2.39 (1.44–3.96) 2.07 (0.84–5.07)
0.45 (0.14–1.51) 1.00 1.04 (0.93–1.14) 1.40 (1.16–1.68) 0.95 (0.59–1.55) 1.93 (0.95–3.94)
0.67 (0.23–1.93) 1.00 1.06 (0.96–1.17) 1.52 (1.27–1.83) 1.17 (0.74–1.87) 2.01 (0.99–4.11)
0.56 (0.08–4.14) 1.00 1.05 (0.89–1.24) 1.63 (1.24–2.15) 0.61 (0.22–1.66) 4.51 (2.02–10.08)
1.17 (0.54–2.54) 1.00 1.04 (0.95–1.13) 1.29 (1.08–1.53) 1.15 (0.76–1.74) 1.03 (0.47–2.27)
2.01 (1.76–2.29) 0.88 (0.83–0.93)
1.16 (1.05–1.28) 0.91 (0.87–0.95)
1.32 (1.19–1.46) 0.90 (0.86–0.93)
1.18 (1.00–1.39) 0.97 (0.91–1.04)
1.32 (1.20–1.44) 0.96 (0.93–0.99)
Note: Values are expressed as odds ratio (95% confidence interval). One multivariate model per semen parameter. BMI ¼ body mass index. Belloc. Obesity affects semen production. Fertil Steril 2014.
Our results conflict with those published by Relwani et al. (4), who did not find any impact of BMI on semen characteristics. They also contradict the recent meta-analysis (5), but it can be noted that those authors found a negative relationship for T, SHBG, and free T (21, 22) with increased BMI, which may, at least in part, explain our findings. Similar findings were reported by Al-Ali et al. (23), with no influence on all assessed sperm quality parameters, but on LH (P¼ .001), T (P< .001), and PRL (P¼ .044). In a cohort of Danish men, serum T, SHBG, and inhibin B all decreased with increasing BMI, whereas free androgen index and E2 increased with increasing BMI. Serum FSH was higher among slim men (24). One could speculate that the effect of obesity on clinical outcomes may be mediated through semen quality. However, the Colaci et al. analysis (7) indicated that the relationship between male obesity and live birth rates was independent from sperm parameters. No data suggest any plausible biologic explanation for the higher fertilization rate and higher live birth observed with assisted reproductive technologies among couples with an obese man in that study. The limited sample size, and the potential effect of other confounders, such as female age, have to be examined in future studies. Interestingly, Ramlau-Hansen et al. in 2007 (25) published a study in which the risk of subfecundity (time to achieve pregnancy >12 months) was estimated in 47,835 pregnancies according to the body weight of both members of the couple. The OR for subfecundity increased between underweight and obese men and women in a dose-response manner. Obesity has been related also to a more than twofold risk of sperm DNA damage (OR 2.5, 95% CI 1.2–5.1) on 330 men in infertile couples (26). Another study (27) found a small increase of oxidative stress with increasing BMI, even if the magnitude of this increase was of minor clinical significance, without an associated decline in sperm DNA integrity or sperm motility with increasing reactive oxygen species production. Our results also contradicted those of MacDonald et al. (8), who did not find an effect on concentration or motility of overweight, but a light impact on morphology. However, their study relied on a relatively small number, with a power only able to detect at least a doubling in abnormally low 1272
sperm concentration and total sperm count (OR R2.0) for overweight or obese men compared with men with a normal BMI. One limitation of our study is that it was conducted among men seeking fertility treatment. Moreover, to improve our knowledge about the effect of weight or obese status on reproduction, researchers should always register for both men and women not only their BMI but also factors associated with weight excess that could also affect fertility, such as lifestyle habits, fat distribution, and associated pathologies. For example, Hammiche et al. (16) found that not only BMI, but also waist circumference R102 cm, a measure for central adiposity, was inversely associated with sperm concentration and total motile sperm count (P< .001 and P< .02, respectively). This was confirmed by a recent paper on 501 men (28). It is also interesting to note that H akonsen et al. (22), on a relatively low sample size, concluded that weight loss could improve semen quality. How increased BMI can alter semen quality has been the subject of a number of studies in nonhuman animal models. A recent study (29) investigated the effect of diet-induced paternal obesity, in the absence of diabetes, on the metabolic health of two resultant generations of mice and concluded that it can modulate sperm microRNA content and germ cell methylation status. Similarly, a study on rats (30) stated, ‘‘observations demonstrate [that] hydrocarbons can promote epigenetic transgenerational inheritance of disease and sperm epimutations, potential biomarkers for ancestral exposure.’’ Induced obesity may also lead to decreased sperm reserves and acceleration of transit time in the epididymis of adult male rats (30). Conversely, diet and exercise may reverse perturbed sperm function (31), and Palmer et al. (32) proposed a role for mammalian SIRT6 protein in spermatogenesis, which is known to be compromised by male obesity. Although it is difficult to generalize our results to the overall population, our results suggest that increased BMI has a deleterious effect on semen concentration but not necessarily on reproductive outcome. Other studies are needed to analyze the benefit of weight reduction among obese men. VOL. 102 NO. 5 / NOVEMBER 2014
Fertility and Sterility® Acknowledgments: The authors thank Denny Sakkas for assistance in reviewing the manuscript.
REFERENCES 1.
2. 3.
4.
5.
6.
7.
8.
9. 10. 11.
12. 13.
14.
15. 16.
17.
Balen AH, Anderson RA, Policy and Practice Committee of the BFS. Impact of obesity on female reproductive health: British Fertility Society, policy and practice guidelines. Hum Fertil 2007;10:195–206. Pasquali R, Patton L, Gambineri A. Obesity and infertility. Curr Opin Endocrinol Diabetes Obes 2007;14:482–7. Maheshwari A, Stofberg L, Bhattacharya S. Effect of overweight and obesity on assisted reproductive technology—a systematic review. Hum Reprod Update 2007;13:433–44. Relwani R, Berger D, Santoro N, Hickmon C, Nihsen M, Zapantis A, et al. Semen parameters are unrelated to BMI but vary with SSRI use and prior urological surgery. Reprod Sci 2011;18:391–7. MacDonald AA, Herbison GP, Showell M, Farquhar CM. The impact of body mass index on semen parameters and reproductive hormones in human males: a systematic review with meta-analysis. Hum Reprod Update 2010; 16:293–311. Shayeb AG, Harrild K, Mathers E, Bhattacharya S. An exploration of the association between male body mass index and semen quality. Reprod Biomed Online 2011;23:717–23. Colaci DS, Afeiche M, Gaskins AJ, Wright DL, Toth TL, Tanrikut C, et al. Men's body mass index in relation to embryo quality and clinical outcomes in couples undergoing in vitro fertilization. Fertil Steril 2012;98:1193–9. Macdonald AA, Stewart AW, Farquhar CM. Body mass index in relation to semen quality and reproductive hormones in New Zealand men: a crosssectional study in fertility clinics. Hum Reprod 2013;28:3178–87. Barazani Y, Katz BF, Nagler HM, Stember DS. Lifestyle, environment, and male reproductive health. Urol Clin North Am 2014;41:55–66. Palmer NO, Bakos HW, Fullston T, Lane M. Impact of obesity on male fertility, sperm function and molecular composition. Spermatogenesis 2012;2:1–11. Sermondade N, Faure C, Fezeu L, Shaye AG, Bonde JB, Jensen TK, et al. BMI in relation to sperm count: an updated systematic review and collaborative meta-analysis. Hum Reprod Update 2013;19:221–31. WHO laboratory manual for the examination and processing of human sperm. 5th ed. Geneva: World Health Organization; 2010. Auger J, Eustache F. Standardization of the morphological assessment of human spermatozoa according to modified David's classification. Andrologie 2000;10:353–87. Blanchard M, Haguenoer K, Apert A, Poret H, Barthelemy C, Royere D, et al. Sperm morphology assessment using David's classification: time to switch to strict criteria? Prospective comparative analysis in a selected IVF population. Int J Androl 2011;2:145–52. World Health Organisation. BMI classifications. Available from: www.who. int/bmi/index.jsp?introPage¼intro_3.html; 2006. Hammiche F, Laven JS, Twigt JM, Boellaard WP, Steegers EA, SteegersTheunissen RP. Body mass index and central adiposity are associated with sperm quality in men of subfertile couples. Hum Reprod 2012;27:2365–72. Cooper TG, Noonan E, von Eckardstein S, Auger J, Baker HW, Behre HM, et al. World Health Organization reference values for human semen characteristics. Hum Reprod Update 2010;16:231–45.
VOL. 102 NO. 5 / NOVEMBER 2014
18.
19.
20.
21.
22.
23.
24.
25.
26.
27. 28.
29.
30.
31.
32.
Prisant N, Cohen-Bacrie P, Amar E, Belaisch-Allart J, Cohen-Bacrie M, Olivennes F, et al. Teratozoospermia, myth or reality? A 10-years retrospective study on 101404 consecutive sperm samples. Gynecol Obstet Fertil 2011;39:136–40. Rolland M, le Moal J, Wagner V, Royere D, de Mouzon J. Decline in semen concentration and morphology in a sample of 26609 men close to general population between 1989 and 2005 in France. Hum Reprod 2013;28:462–70. Chavarro JE, Toth TL, Wright DL, Meeker JD, Hauser R. Body mass index in relation to semen quality, sperm DNA integrity, and serum reproductive hormone levels among men attending an infertility clinic. Fertil Steril 2010;93:2222–31. Zumoff B, Strain GW, Miller LK, Rosner W, Senie R, Seres DS, et al. Plasma free and nonsex-hormone-binding-globulin-bound testosterone are decreased in obese men in proportion to their degree of obesity. J Clin Endocrinol Metab 1990;71:929–31. H akonsen LB, Thulstrup AM, Aggerholm AS, Olsen J, Bonde JP, Andersen CY, et al. Does weight loss improve semen quality and reproductive hormones? Results from a cohort of severely obese men. Reprod Health 2011;8:24. M Al-Ali B, Gutschi T, Pummer K, Zigeuner R, Brookman-May S, Wieland WF, et al. Body mass index has no impact on sperm quality but on reproductive hormones levels. Andrologia 2014;46:106–11. Jensen TK, Andersson AM, Jørgensen N, Andersen AG, Carlsen E, Petersen JH, et al. Body mass index in relation to semen quality and reproductive hormones among 1,558 Danish men. Fertil Steril 2004;82:863–70. Ramlau-Hansen CH, Thulstrup AM, Nohr EA, Bonde JP, Sorensen TI, Olsen J. Subfecundity in overweight and obese couples. Hum Reprod 2007;22: 1634–7. Dupont C, Faure C, Sermondade N, Boubaya M, Eustache F, Clement P, et al. Obesity leads to higher risk of sperm DNA damage in infertile patients. Asian J Androl 2013;15:622–5. Tunc O, Bakos HW, Tremellen K. Impact of body mass index on seminal oxidative stress. Andrologia 2011;43:121–8. Eisenberg ML, Kim S, Chen Z, Sundaram R, Schisterman EF, Buck Louis GM. The relationship between male BMI and waist circumference on semen quality: data from the LIFE study. Hum Reprod 2014;29:193–200. Fullston T, Ohlsson Teague EM, Palmer NO, DeBlasio MJ, Mitchell M, Corbett M, et al. Paternal obesity initiates metabolic disturbances in two generations of mice with incomplete penetrance to the F2 generation and alters the transcriptional profile of testis and sperm microRNA content. FASEB J 2013;27:4226–43. Tracey R, Manikkam M, Guerrero-Bosagna C, Skinner MK. Hydrocarbons (jet fuel JP-8) induce epigenetic transgenerational inheritance of obesity, reproductive disease and sperm epimutations. Reprod Toxicol 2013;36: 104–16. Palmer NO, Bakos HW, Owens JA, Setchell BP, Lane M. Diet and exercise in an obese mouse fed a high-fat diet improve metabolic health and reverse perturbed sperm function. Am J Physiol Endocrinol Metab 2012;302: E768–80. Palmer NO, Fullston T, Mitchell M, Setchell BP, Lane M. SIRT6 in mouse spermatogenesis is modulated by diet-induced obesity. Reprod Fertil Dev 2011; 23:929–39.
1273