The association of age and semen quality in healthy men

B. Eskenazi*,1,5, A.J. Wyrobek*,2, E. Sloter2,3, S.A. Kidd1, L. Moore1, S. Young1 and D. Moore2,4

1 School of Public Health, University of California, Berkeley, CA, 2 Biology and Biotechnology Research Program, Lawrence Livermore National Laboratory, Livermore, CA, 3 Genetics and Developmental Biology Program, West Virginia University, Morgantown, WV and 4 California Pacific Medical Center, San Francisco, CA, USA

5 To whom correspondence should be addressed at: 140 Earl Warren Hall, School of Public Health, University of California, Berkeley, CA 94720-7360, USA. e-mail: eskenazi{at}uclink4.berkeley.edu


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Although the effect of maternal age on fertility is well known, it is unclear whether paternal age also affects fertility. This cross-sectional study sought to characterize the association between age and semen quality, a well-known proxy of fertility status. METHODS: A convenience sample of 97 non-smoking men (aged 22–80 years) without known fertility problems was recruited from a national government laboratory. The men provided semen samples and information relating to lifestyle, diet, medical and occupational details. Semen volume (ml), sperm concentration (x106/ml), total sperm count (x106), motility (%), progressive motility (%) and total progressively motile sperm count (x106) were measured. RESULTS: After adjusting for covariates, semen volume decreased by 0.03 ml per year of age (95% CI: –0.05, –0.01); motility decreased by 0.7% per year (95% CI: –0.92, –0.43); progressive motility decreased by 3.1% per year (95% CI: –4.5, –1.6); and total progressively motile sperm count decreased by 4.7% per year (95% CI: –7.2, –2.2). There was a suggested decrease in sperm concentration and count. The proportion of men with abnormal volume, concentration and motility was significantly increased across the age decades. CONCLUSIONS: In a convenience sample of healthy men from a non-clinical setting, semen volume and sperm motility decreased continuously between 22–80 years of age, with no evidence of a threshold.

Key words: age/semen volume/sperm concentration/sperm motility/total sperm count


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Approximately 15% of couples of reproductive age experience infertility, and more than a quarter of infertility cases may be attributed to male factors (Templeton, 1995Go). It is well known that maternal age is a significant contributor to human infertility (Joffe and Li, 1994Go), due primarily to the precipitous loss of functional oocytes in women by their late thirties (Tietze, 1957Go; Lansac, 1995Go). Human spermatogenesis, on the other hand, continues well into advanced ages, allowing men to reproduce during senescence. Although very little is known about the topic, paternal age may also contribute to human infertility.

Understanding the effect of male age on fertility has become increasingly important in public health because a growing number of men are choosing to father children at older ages (Ventura et al., 1997Go). In the United States, for example, there has been a 24% increase in the birth rate for fathers aged 35 to 54 years since 1980. However, advanced male age has been associated with significant reductions in pregnancy rates, increased time-to-pregnancy and increased subfecundity (Kidd et al., 2001Go). In a recent study of 8515 planned pregnancies (of greater than 24 weeks gestation), men older than 35 years had half the chance of fathering a child within 12 months compared with men aged less than 25 years, after controlling for women’s age and other factors (Ford et al., 2000Go).

Semen quality is generally considered to be a proxy measure of male fertility, and changes in semen quality can occur after exposure to toxic agents (Wyrobek et al., 1983Go; Wyrobek, 1993Go) or from host factor effects such as age (Kidd et al., 2001Go). The weight of evidence primarily from clinical studies (see Kidd et al., 2001 for a review) suggests that age is associated with diminished semen volume, sperm motility and/or sperm morphology, but that sperm concentration is affected little by age (Schwartz et al., 1983Go; Wang et al., 1985Go; Abramsson, 1988Go; Carlsen et al., 1992Go; Berling and Wolner-Hanssen, 1997Go; Lemcke et al., 1997Go; Spandorfer et al., 1998Go). However, it is unclear whether these observations are applicable to the general population of healthy men. Also, men at older ages (e.g. >50 years) were under-represented in many of these clinical studies, which limited statistical power and prevented the determination of the shape of the relationship between age and semen quality. In addition, potential confounders that might explain changes with age, such as smoking history or duration of abstinence, were seldom controlled.

The purpose of the current research was to examine the magnitude and the shape of the relationships between age and semen volume, sperm concentration and sperm motility in a non-clinical group of men selected to span across a wide range of age, from their 20s to their 80s. These men provided extensive information on medical, lifestyle and occupational exposure factors that could affect semen quality.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Subjects
The study population of the Age and Genetic Effects on Sperm (AGES) Study consisted of a convenience sample of 97 male volunteers employed or retired from the Lawrence Livermore National Laboratory (LLNL) located in Livermore, California. LLNL was used as the recruitment site because the workforce is relatively homogeneous and because it was the site of the semen analysis laboratory. The AGES Study was approved by the Institutional Review Boards of the participating institutions and all subjects gave written informed consent.

Male workers and retirees were recruited between October 1997 and July 1998 from advertisements posted at the LLNL, e-mail listings, and in newsletters. Preliminary screening over the telephone excluded men who had smoked cigarettes in the last 6 months; had a vasectomy or a history of an undescended testicle or prostate cancer; had ever received chemotherapy or radiation treatments; had current fertility or reproductive problems (men were included who claimed their wives had a fertility problem); or had a previous semen analysis with zero sperm count. Twenty men were excluded after the preliminary screening due to current smoking (n = 11), varicocele (n = 3), one testicle (n = 2), undescended testicles (n = 1), valium use (n = 1), chemotherapy (n = 1) and hepatitis B infection (n = 1). At least 15 men were recruited from each age decade from age 20 to age 70 years; additional men above age 70 were also enrolled.

Eligible men were mailed a questionnaire, a semen collection container, and a small foam-insulated vacuum container (Aladdin Industries Inc., Nashville, TN, USA). The men were instructed to provide a semen sample by masturbation into the collection container after abstaining from ejaculating for 2–7 days, and to deliver it inside the insulated container to a drop box at LLNL. The questionnaire gathered information on medical history, reproductive history (e.g. fatherhood history), sociodemographic characteristics (age, race, education) and lifestyle habits (e.g. diet, alcohol, cigarette and caffeine consumption). Completed questionnaires were mailed to the University of California at Berkeley. The completed questionnaires were reviewed with the participant over the telephone.

Semen analysis
All semen samples were delivered within 2 h of collection. One laboratory technician performed all the analyses, and was blinded to the age and identity of the donors. Semen analysis was performed immediately upon receipt of the sample, averaging about 45 min after its production (range 5–120 min). Duplicate samples were requested from donors if the sample exhibited any of the following: volume <1 ml, zero motility, abnormal numbers of red or white blood cells, or potential loss of sample reported by donor. In total, 115 fresh semen specimens were provided from 97 men. The study results did not change regardless of whether a man’s first or second sample was used for the analyses.

The semen analysis was performed using slightly modified protocols provided by P. Turek of the University of California, San Francisco, Department of Urology and by Hamilton Thorne Research (Beverly, MA, USA), following published guidelines (World Health Organization, 1992). Each specimen was liquefied at room temperature and the volume measured to the nearest 0.1 ml. Sperm concentration of gently mixed samples was determined for three independent loadings on a Neubauer haemocytometer. For sperm motility analysis, 50 µl of semen were diluted to a standardized sperm concentration of ~35x106 sperm per ml using Dulbecco’s phosphate-buffered saline solution supplemented with glucose (1 g/l) and bovine serum albumin (World Health Organization, 1992). At least 150 motile and immotile sperm were evaluated at 400x phase-contrast magnification using a 2X-CEL 20 µm depth chambered microscope slide (Hamilton Thorne Research) maintained at 37°C by a MiniTherm slide warmer (Hamilton Thorne Research). If >10% difference was found between the two analyses of the same sample, the sample was re-analysed. All cell counting was performed on a Macintosh computer using a modified version of CytoScore© developed at LLNL.

Statistical analysis
All analyses were conducted using STATA 7.0 (Stata Statistical Software, 1999). Total sperm count was calculated by multiplying semen volume by sperm concentration. Total progressively motile sperm was defined as the product of total sperm count and percent progressive motility. Volume and motility appeared to be normally distributed, while sperm progressive motility, concentration, total count, and total progressively motile sperm were log-transformed to achieve normality. Zero values for semen parameters were re-coded to half of the lowest measured value before transformation. All count and concentration analyses were performed with and without four subjects with azoospermia; all motility analyses were performed without these men.

Separate multiple linear regression models were used to examine the relationship of age with each semen parameter controlling for the following potential confounding factors: duration of abstinence; average ejaculation frequency (last 3 months); time from sample collection to sample processing; season; time worked at LLNL; body mass index; ethnicity; years of education; potential occupational exposures (last 3 months); dosimetry records; time spent sitting; medication use; alcohol and caffeine intake; history of tobacco use; history of chronic disease (e.g. high blood pressure, heart problems or diabetes); history of genitourinary disease (e.g. urinary tract or other genitourinary infection, sexually transmitted disease or past history of infertility); and fatherhood history. Covariates were included in the multivariate models if they changed the regression coefficient for age by >10% or if they were statistically significant at P < 0.1.

Results for log-transformed semen parameters were presented as the relative percent change per year as converted from the antilog of the regression coefficient. Results for untransformed semen parameters were presented as the absolute change per year of age as well as the relative change in the outcome in men who were age 50 years compared with age 30 years. The partial correlation for age, adjusted for the other covariates in the model, is presented.

A hockey-stick model was fitted to the adjusted data to determine if there was a change in slope at any age (Bacon and Watts, 1971Go). Hockey-stick analysis of the adjusted regression models did not fit the data better than the log transformations based on a likelihood test, ruling out any clear ‘threshold’ effect for any of the semen parameters. Thus, only the results of the regression analyses are presented.

Maximum-likelihood logit models were fit to the data for each semen parameter to estimate the probability of having an abnormal value at each year of age, adjusting for the same covariates that were in the linear regression models. Abnormal semen values were defined from WHO (1992) standards: volume <=2 ml, concentration <20x106/ml, count <=40x106, motility <50%, progressive motility <25%, and total progressively motile sperm count <10x106.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Characteristics of study population
The 97 men were on average 46.4 (±15.8, SD) years old (range 22–80 years). The group was 91% Caucasian, 80% college graduates and 72% never-users of tobacco. Ethnicity and level of education did not vary by age of the man (data not shown). As shown in Table I, older age was associated with prior tobacco use (none smoked cigarettes during the previous 6 months), increased duration of sexual abstinence, and ever having had high blood pressure. Age was unrelated to time between collection and sample processing, alcohol use, body mass or history of urinary tract infection. Some 64% of the men had fathered children earlier in life; and as expected, men with proven fertility were older than men with unproven fertility (50.6 versus 39.0 years; P = 0.001; data not shown).


View this table:
[in this window]
[in a new window]
 
Table I. Mean (SD) of age and semen parameters by selected characteristics of the participants
 
The median semen volume was 2.7 ml; median sperm concentration was 93x106/ml; and median total sperm count was 259x106 (Table II). The median motility parameters were 42% motility, 20% progressive motility and 55x106 progressively motile sperm. Seven men, all of whom were aged over 60 years, were either azoospermic (63, 77, 77 and 78 years old) or had no progressively motile sperm (60, 64 and 67 years old). Two of the four azoospermic men and all of the men with zero sperm motility had fathered children earlier in their life, and none had ever been diagnosed with their own fertility problem. The aetiology of the azoospermia was unknown.


View this table:
[in this window]
[in a new window]
 
Table II. Median and interquartile range (IQR, 25th–75th percentile) of semen parameters by age decade
 
Semen quality varied with some lifestyle and medical factors (Table I). Semen volume was higher in men who never used tobacco or who never drank alcohol. Sperm concentration was higher in those who had abstained from ejaculation for more than 5 days. Sperm count was also higher in men who had abstained for >5 days, but lower in men who had ever had high blood pressure. Both percent motile and percent progressively motile sperm were lower in men who had ever had a urinary tract infection. Total number of progressively motile sperm was lower in men who had ever had high blood pressure.

Semen volume
Men in their 20s had a median semen volume of 3.0 ml (Table II), and there was a significant trend toward lower volumes across age decades (P-value for trend, < 0.01). This decrease with age is shown graphically in Figure 1A. In the multiple linear regression analyses, semen volume decreased with age by 0.03 ml per year (95% CI = –0.05, –0.02) after adjusting for length of abstinence and prior use of tobacco and alcohol (Table III). From the multivariate regression models, a 50-year-old man was calculated to have a 20% relative decrease in semen volume compared with a 30-year-old man. Age explained 14.4% (partial r = –0.38, P < 0.001) of the total variance in semen volume. The number of men with abnormal semen volume (<=2 ml) also significantly increased across the age decades (P- value for trend, < 0.001; Figure 2). As shown in the adjusted logit graph (Figure 3), at age 30 years, about 10% of the men would have an abnormal semen volume, and this proportion increased to approximately 35% at age 50 and 80% by age 80 (P < 0.001).



View larger version (23K):
[in this window]
[in a new window]
 
Figure 1. The relationship between age in years and semen volume (A), concentration (B), count (C), sperm motility (D), progressive motility (E) and total progressively motile sperm (F). Individual data points are shown as well as locally weighted smoothed fits (using a bandwidth of 0.8) (broken lines) and unadjusted linear regression lines (solid lines). Detailed results and additional fits adjusted for covariates are described in the text.

 

View this table:
[in this window]
[in a new window]
 
Table III. Unadjusted and adjusted linear regression analyses of age on semen quality
 


View larger version (37K):
[in this window]
[in a new window]
 
Figure 2. The frequency of men in each age decade with abnormal semen values as defined by WHO guidelines. The WHO guidelines used were: volume <2 ml, concentration <20x106/ml, count <40x106, motility <50%, progressive motility <25% and total progressively motile sperm count <10x106.

 


View larger version (17K):
[in this window]
[in a new window]
 
Figure 3. Estimated probability of having a clinically abnormal semen parameter at each year of age using maximum-likelihood adjusted logit graphs. Each logit model was adjusted for the same covariates as the linear regression models. For graphing purposes, all covariates were set to the mean value of the population.

 
Sperm concentration and total sperm count per specimen
Men in their 20s had median sperm concentrations and total sperm counts of 92x106/ml and 345x106 respectively. Total count decreased significantly across age decades (P-value for trend, 0.01) (Table II; Figure 1B). The relationship between age and concentration was statistically significant in the model including the four azoospermic men (slope = –2.5% per year; 95% CI = –4.2, –0.8) (Table III), but non-significant in the model with the four men excluded (slope = –0.6% per year; 95% CI = –2.0, 0.9). The relationship between total sperm count and age was statistically significant, both including the four azoospermic men (slope = –3.5% per year; 95% CI = –5.3, –1.7) and excluding them (slope = –1.7% per year; 95% CI = –3.4, –0.006), although the effect was greatly decreased for the latter.

As shown in Figure 2, the number of men with abnormal levels for sperm concentration and count significantly increased across age decades (P-values for trend 0.01 and 0.003 respectively); however, the number of men affected in each decade was small. Based on the adjusted logit model, the predicted probability of having an abnormal value for concentration was approximately 5% at age 30 years, 10% at age 50 and 35% by age 80 (P = 0.006). For sperm count, the predicted probability for an abnormal value was about 5% at age 30 years, 15% at age 50, and 50% by age 80 (P = 0.03) (Figure 3).

Percent motility, percent progressive motility and total number of progressively motile sperm
Semen specimens provided by men in their 20s had medians of 50.0% motility, 29.0% progressive motility and 96.6x106 total progressively motile sperm (Table II). There were significant trends towards reduced sperm motility across age decades for all three parameters (P-values for trend < 0.01, for all three) (Table I; Figure 1D, E and F). In the regression analyses, percent motility decreased by 0.7% per year of age (95% CI = –0.92, –0.43) after controlling for length of abstinence and time before the sample was processed (Table III). A 50-year-old man was calculated to have a relative decrease of 28% motile sperm compared with a 30-year-old man. Similarly, after adjusting for covariates, there was 3.1% (95% CI = –4.5, –1.6) relative decrease in the percent of progressive motility per year, and a 4.7% decrease per year (95% CI = –7.2, –2.2) for total progressively motile sperm. The proportion of the total variance explained by age was 27.0% for motility (r = –0.52, P < 0.001), 17.6% for log progressive motility (r = –0.42, P < 0.001) and 13.0% for log total progressively motile (r = –0.36, P < 0.001).

The number of men with abnormal percent motile and total progressively motile sperm also increased significantly across age decades (P-value for trend 0.001 and 0.006 respectively). The frequency of percent abnormal progressive motility did not significantly increase across age decades (P-value for trend 0.12), although the frequency appeared to increase up to the 40s and then plateau (Figure 2). The adjusted logit model predicted that about 40% of the men by age 30 years would have abnormal percent motility. This increased to approximately 80% by age 50, and to 100% by age 80 (P < 0.001) (Figure 3). The probability for having abnormal progressive motility was approximately 52% at age 30 years, 67% at age 50, and 84% by age 80 (P = 0.07). The probability for having abnormal total progressively motile sperm was approximately 5% at age 30 years, 15% at age 50, and 50% by age 80 (P = 0.01).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
This investigation found evidence of significant age-dependent reductions in several aspects of semen quality among a cohort of healthy workers and retirees: the largest effects of age were on sperm motility, with intermediate effects on semen volume and the smallest effects on sperm numbers. The age effects on sperm concentration and total count appeared to be driven by four azoospermic men over age 60 years. For men aged 50 years, an approximate 80% probability of clinically abnormal motility, 35% probability of abnormally low semen volume and 15% probability of abnormally low sperm count was predicted. These probabilities increased to approximately 100, 80 and 50% respectively for men aged 80 years. There was no evidence of an age ‘threshold" for any of the semen parameters, but rather a gradual change over time.

These findings support and extend those for semen quality in prior clinical studies of infertile patients and sperm donors (Kidd et al., 2001Go). However, the percent changes with age in the present study were generally at the high end of changes observed in the clinical studies. For example, a 20% decrease was observed in semen volume in 50-year-old compared to 30-year-old men, which is near the high end of the 3 to 22% range reported in prior clinical studies (Fisch et al., 1996Go; Rolf et al., 1996Go; Andolz et al., 1999Go; Kidd et al., 2001Go). Similarly, the 28% relative decrease in motility identified was near the high end of the 3 to 37% range of decreases reported in clinical studies (Schwartz et al., 1983Go; Auger et al., 1995Go; Fisch et al., 1996Go; Rolf et al., 1996Go). Sperm concentration was least affected by age among the semen quality parameters studied, although there was a significant trend toward increased numbers of men with abnormal concentration across age decades. This was not inconsistent with reports from the clinical literature which, in general, did not find that sperm concentration decreased with age (Schwartz et al., 1983Go; Wang et al., 1985Go; Abramsson, 1988Go; Carlsen et al., 1992Go; Berling and Wolner-Hanssen, 1997Go; Lemcke et al., 1997Go; Spandorfer et al., 1998Go; Kidd et al., 2001Go). Although the present age-related findings for total sperm count were more robust, this was likely driven by the strong association of age with semen volume.

The present findings are generally consistent with some previous studies which demonstrated a decreased pregnancy rate and longer time to pregnancy in partners of older men (Stanwell-Smith and Hendry, 1984Go; Abramsson, 1988Go; Ducot et al., 1988Go; Goldman and Montgomery, 1989Go; Olsen, 1990Go; Mathieu et al., 1995Go; Rolf et al., 1996Go; Brzechffa et al., 1998Go; Spandorfer et al., 1998Go), but not others (Abramsson, 1988Go; Gallardo et al., 1996Go; van der Westerlaken et al., 1998Go; Paulson et al., 2001Go). Many of the pregnancy-based studies may be limited because they do not control for maternal age (Stanwell-Smith and Hendry, 1984Go; Bostofte, 1987Go; Abramsson, 1988Go; Ducot et al., 1988Go; Ford et al., 1994Go; Joffe and Li, 1994Go; van der Westerlaken et al., 1998Go) and/or they may not include sufficient numbers of men in the older age ranges (Bostofte, 1987Go; Abramsson, 1988Go; Ducot et al., 1988Go; Olsen, 1990Go; Ford et al., 1994Go; Mathieu et al., 1995Go; Gallardo et al., 1996Go; Rolf et al., 1996Go; Spandorfer et al., 1998Go; Paulson et al., 2001Go).

The design of the present study has several notable strengths. The ability to detect an association between age and semen parameters was enhanced by the inclusion of relatively larger numbers of older men compared with most previous studies (Wang et al., 1985Go; Abramsson, 1988Go; Check et al., 1989Go; Gallardo et al., 1996Go; Irvine et al., 1996Go; Spandorfer et al., 1998Go). The finding of decreased sperm motility with age was consistent with the findings of others (Nieschlag et al., 1982Go), whose study also included a large number of men aged over 60 years. Also, by including men over a wide age range, the shape of the relationship between age and semen quality could be examined. In addition, the men in this study were relatively homogeneous in sociodemographic characteristics across age groups. The sample consisted of employed and retired workers from a single occupational setting, of middle to high socioeconomic class, and with employer-paid access to medical care. Unlike most previous studies, control was introduced for other demographic or lifestyle differences in the statistical analysis (Homonnai et al., 1982Go; Nieschlag et al., 1982Go; Schwartz et al., 1983Go; Dondero et al., 1985Go; Wang et al., 1985Go; Abramsson, 1988Go; Check et al., 1989Go; Singer et al., 1990Go; Carlsen et al., 1992Go; Mladenovic, 1994Go; Bujan et al., 1996Go; Gallardo et al., 1996Go; Haidl et al., 1996Go; Irvine et al., 1996Go; Rolf et al., 1996Go; Berling and Wolner-Hanssen, 1997Go; Lemcke et al., 1997Go; Spandorfer et al., 1998Go). However, as in all semen quality studies, the present sample included only those men who were willing to provide semen specimens and thus, may not have been representative of the general population of sexually active men. Several different sources of bias can be introduced by a non-representative sample. Although no men with known fertility problems volunteered to participate, men who suspected a potential fertility problem may have been more likely to volunteer. If this differed by age, bias could have been introduced, although there was no evidence for this. In addition, since the volunteers were all workers or retirees, a ‘healthy worker’ bias could have been introduced into the study, resulting in an underestimate of the age effect.

At least two broad modes of action may explain the age-dependent changes observed in semen quality. First, there may be cellular or physiological changes in the genitourinary tract with ageing. In autopsies of men who died from accidental causes, there have been age-related narrowing and sclerosis of the testicular tubular lumen, decreases in spermatogenic activity, increased degeneration of germ cells, and decreased numbers and function of Leydig cells (Bishop, 1970Go; Johnson, 1986Go). Decreased semen volume with age may be caused by seminal vesicle insufficiency, since seminal vesicle fluid contributes most of the ejaculate volume (Hamilton and Naftolin, 1981Go; Goldman and Montgomery, 1989Go). Changes in the prostate that occur with ageing, such as smooth muscle atrophy and a decrease in protein and water content, may contribute to decreased semen volume and sperm motility (Schneider, 1978Go). In addition, there may be age-related changes in the epididymis where sperm acquire the capacity for vigorous forward motility during transit. The epididymis is a hormonally sensitive tissue, which plays an important role in sperm maturation (Hamilton and Naftolin, 1981Go). Thus, hormonal or epididymal senescence may lead to decreased motility in older men. Also, older men may have decreased capacity to repair cellular and tissue damage from toxicant or disease exposure.

Second, age provides increased opportunities to suffer reproductive damage from exogenous exposures or diseases (Wyrobek et al., 1983Go; Wyrobek, 1993Go). Older men are more likely to have smoked and to have smoked for a longer period than younger men, or to have had illnesses including genitourinary infections. Male age may also be a proxy for a ‘cohort effect’; that is, a common specific exposure experienced by men in the same birth cohort. For example, men who were born prior to 1972 were more likely to have been exposed to DDT, an endocrine disruptor, which was later banned (Carlsen et al., 1992Go; Kidd et al., 2001Go). In the present study, the decline in semen quality could be due to some unknown occupational exposure that was related to age, or time worked at LLNL. However, the analyses provided no evidence that time worked at LLNL or any occupational exposure explained the age-related decline in either sperm count or motility.

The semen parameters evaluated herein are not expected to be the only sperm end-points that will show age-related damage. Other parameters that may be affected by age include sperm morphology, which has been shown to be a sensitive indicator of the status of the germinal epithelium (MacLeod, 1964Go; Wyrobek, 1983Go). Several studies have suggested age-related defects in the genetic integrity of the sperm. For example, age has been associated with increased sperm aneuploidy in humans (Griffin et al., 1995Go; Martin et al., 1995Go; Robbins et al., 1995Go; Lowe et al., 2001Go) and in mice (Lowe et al., 1995Go), and with increases in the incidence of de-novo germinal mutations (Friedman, 1981Go; Martinez-Frias et al., 1988Go; Modell and Kuliev, 1990Go).

In summary, significant age-related decreases in semen quality were observed, most notably for semen volume and sperm motility. Because semen quality is a proxy for fertility, these data suggest that men may become progressively less fertile as they age. However, unlike women, there appears to be no evidence of an age threshold for men. The present findings have important implications for men who choose to delay fatherhood, since they may reduce their chance of success the longer they delay.


    Acknowledgements
 
The authors thank Xiu Lowe, M.D., for assistance in the development of laboratory methods and quality control procedures, and Elana Wallenstein for assisting in the statistical analyses. They also thank Paul Turek, M.D., at the University of California, San Francisco, for training and guidance in the development of the laboratory procedures for sperm analyses; and Joginder Nath, Ph.D., at West Virginia University for his financial and research support of E.S. This publication was made possible by grant number P42 ES04705 from the National Institute of Environmental Health Sciences, NIH, with funding provided by EPA; and was performed in part under the auspices of the US DOE by LLNL, under contract W-405-ENG-48. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of the NIEHS, NIH, or EPA.


    FOOTNOTES
 
* B.Eskenazi and A.J.Wyrobek were equally responsible for this work Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Abramsson, L. (1988) On the investigation of men from infertile relations: A clinical study with special regard to anamnesis, physical examination, semen-, hormone-, and chromosome analyses, from men with non-‘normal’ semen. Scand. J. Urol. Nephrol. Suppl., 113, 7–47.

Andolz, P., Bielsa, M.A. and Vila, J. (1999) Evolution of semen quality in North-eastern Spain: a study in 22,759 infertile men over a 36 year period. Hum. Reprod., 14, 731–735.[Abstract/Free Full Text]

Auger, J., Kunstmann, J.M., Czyglik, F. and Jouannet, P. (1995) Decline in semen quality among fertile men in Paris during the past 20 years [see comments]. N. Engl. J. Med., 332, 281–285.[Abstract/Free Full Text]

Bacon, D. and Watts, D. (1971) Estimating the transition between two intersecting straight lines. Biometrika, 58, 525–534.[ISI]

Berling, S. and Wolner-Hanssen, P. (1997) No evidence of deteriorating semen quality among men in infertile relationships during the last decade: a study of males from Southern Sweden. Hum. Reprod., 12, 1002–1005.[CrossRef][ISI][Medline]

Bishop, M.W. (1970) Ageing and reproduction in the male. J. Reprod. Fertil. Suppl., 12, 65–87.[Medline]

Bostofte, E. (1987) Prognostic parameters in predicting pregnancy. A twenty-year follow-up study comprising semen analysis in 765 men of infertile couples evaluated by the Cox regression model. Acta Obstet. Gynecol. Scand., 66, 617–624.[ISI][Medline]

Brzechffa, P.R., Daneshmand, S. and Buyalos, R.P. (1998) Sequential clomiphene citrate and human menopausal gonadotrophin with intrauterine insemination: the effect of patient age on clinical outcome. Hum. Reprod., 13, 2110–2114.[Abstract]

Bujan, L., Mansat, A., Pontonnier, F. and Mieusset, R. (1996) Time series analysis of sperm concentration in fertile men in Toulouse, France between 1977 and 1992. Br. Med. J., 312, 471–472.[Abstract/Free Full Text]

Carlsen, E., Giwercman, A., Keiding, N. and Skakkebaek, N.E. (1992) Evidence for decreasing quality of semen during past 50 years [see comments]. Br. Med. J., 305, 609–613.[ISI][Medline]

Check, J.H., Shanis, B., Bollendorf, A., Adelson, H. and Breen, E. (1989) Semen characteristics and infertility in aging. Arch. Androl., 23, 275–277.[ISI][Medline]

Dondero, F., Mazzilli, F., Giovenco, P., Lenzi, A. and Cerasaro, M. (1985) Fertility in older men. J. Endocrinol. Invest., 8, 87–91.[Medline]

Ducot, B., Spira, A., Feneux, D. and Jouannet, P. (1988) Male factors and the likelihood of pregnancy in infertile couples. II. Study of clinical characteristics – practical consequences. Int. J. Androl., 11, 395–404.[ISI][Medline]

Fisch, H., Goluboff, E., Olson, J., Feldshuh, J., Broder, S. and Barad, D. (1996) Semen analyses in 1283 men from the United State over a 25-year period: no decline in quality. Fertil. Steril., 65, 1009–1014.[ISI][Medline]

Ford, J., MacCormac, L. and Hiller, J. (1994) PALS (pregnancy and lifestyle study): association between occupational and environmental exposure to chemical and reproductive outcome. Mutat. Res., 313, 153–164.[CrossRef][ISI][Medline]

Ford, W.C., North, K., Taylor, H., Farrow, A., Hull, M.G. and Golding, J. (2000) Increasing paternal age is associated with delayed conception in a large population of fertile couples: evidence for declining fecundity in older men. The ALSPAC Study Team (Avon Longitudinal Study of Pregnancy and Childhood). Hum. Reprod., 15, 1703–1708.[Abstract/Free Full Text]

Friedman, J. (1981) Genetic disease in the offspring of older fathers. Obstet. Gynecol., 57, 745–749.[Abstract]

Gallardo, E., Simón, C., Levy, M., Guanes, P., Remohí, J. and Pellicer, A. (1996) Effect of age on sperm fertility potential: oocyte donation as a model. Fertil. Steril., 66, 260–264.[ISI][Medline]

Goldman, N. and Montgomery, M. (1989) Fecundability and husband’s age. Soc. Biol., 36, 146–166.[ISI][Medline]

Griffin, D., Abruzzo, M., Millie, E., Cheean, L., Feingold, E., Sherman, S. and Hassold, T. (1995) Non-disjunction in human sperm: evidence for an effect of increasing paternal age. Hum. Mol. Genet., 4, 2227–2232.[Abstract]

Haidl, G., Jung, A. and Schill, W.B. (1996) Ageing and sperm function. Hum. Reprod., 11, 558–560.[Abstract]

Hamilton, D. and Naftolin, F. (1981) Basic Reproductive Medicine. MIT Press, Cambridge, UK.

Homonnai, Z.T., Fainman, N., David, M.P. and Paz, G.F. (1982) Semen quality and sex hormone pattern of 29 middle aged men. Andrologia, 14, 164–170.[ISI][Medline]

Irvine, S., Carwood, E., Richardson, D., MacDonald, E. and Aitken, J. (1996) Evidence of deteriorating semen quality in the United Kingdom: birth cohort study in 577 in Scotland over 11 years. Br. Med. J., 312, 467–471.[Abstract/Free Full Text]

Joffe, M. and Li, Z. (1994) Male and female factors in fertility. Am. J. Epidemiol., 140, 921–929.[Abstract]

Johnson, L. (1986) Spermatogenesis and aging in the human. J. Androl., 7, 331–354.[Free Full Text]

Kidd, S.A., Eskenazi, B. and Wyrobek, A.J. (2001) Effects of male age on semen quality and fertility: a review of the literature. Fertil. Steril., 75, 237–248.[CrossRef][ISI][Medline]

Lansac, J. (1995) Delayed parenting. Is delayed childbearing a good thing? Hum. Reprod., 10, 1033–1035.[ISI][Medline]

Lemcke, B., Behre, H.M. and Nieschlag, E. (1997) Frequently subnormal semen profiles of normal volunteers recruited over 17 years. Int. J. Androl., 20, 144–152.[CrossRef][ISI][Medline]

Lowe, X., Collins, B., Allen, J., Holland, N., Breneman, J., van Beek, M., Bishop, J. and Wyrobek, A. (1995) Aneuploidies and micronuclei in the germ cells of male mice of advanced age. Mutat. Res., 338, 59–76.[CrossRef][ISI][Medline]

Lowe, X., Eskenazi, B., Nelson, D.O., Kidd, S., Alme, A. and Wyrobek, A.J. (2001) Frequency of XY sperm increases with age in fathers of boys with Klinefelter syndrome. Am. J. Hum. Genet., 69, 1046–1054.[CrossRef][ISI][Medline]

MacLeod, J. (1964) Human seminal cytology as a sensitive indicator of the germinal epithelium. Int. J. Fertil., 9, 281–295.[ISI]

Martin, R., Spriggs, E., Ko, E. and Rademaker, A. (1995) The relationship between paternal age, sex ratios, and aneuploidy frequencies in human sperm, as assessed by multicolor FISH. Am. J. Hum. Genet., 57, 1395–1399.[Medline]

Martinez-Frias, M., Herranz, I., Salvador, J., Prieto, L., Ramos-Arroyo, M., Rodriguez-Pinilla, E. and Cordero, J. (1988) Prevalence of dominant mutations in Spain: effect of changes in maternal age distribution. Am. J. Hum. Genet., 31, 845–852.

Mathieu, C., Ecochard, R., Bied, V., Lornage, J. and Czyba, J. (1995) Cumulative conception rate following intrauterine artificial insemination with husband’s spermatozoa: influence of husband’s age. Hum. Reprod., 10, 1090–1097.[Abstract]

Mladenovic, I. (1994) Sperm morphology and motility in different age populations. Arch. Androl., 32, 197–205.[ISI][Medline]

Modell, B. and Kuliev, A. (1990) Changing paternal age distribution and the human mutation rate in Europe. Hum. Genet., 86, 198–202.[ISI][Medline]

Nieschlag, E., Lammers, U., Freischem, C., Langer, K. and Wickings, E. (1982) Reproductive functions in young fathers and grandfathers. J. Clin. Endocrinol. Metab., 55, 676–681.[Abstract]

Olsen, J. (1990) Subfertility according to the age of the mother and the father. Dan. Med. Bull., 37, 281–282.[ISI][Medline]

Paulson, R., Milligan, R. and Sokol, R. (2001) The lack of influence of age on male fertility. Am. J. Obstet. Gynecol., 184, 818–822.[CrossRef][ISI][Medline]

Robbins, W., Baulch, J., Moore, D.I., Weire, H.-U., Blakey, D. and Wyrobek, A. (1995) Three-probe fluorescence in situ hybridization to assess chromosome X, Y, and 8 aneuploidy in sperm of 14 men from two healthy groups: evidence for a paternal age effect on sperm aneuploidy. Reprod. Fertil. Dev., 7, 799–809.[ISI][Medline]

Rolf, C., Behre, H. and Nieschlag, E. (1996) Reproductive parameters of older compared to younger men of infertile couples. Int. J. Androl., 19, 135–142.[ISI][Medline]

Schneider, E.L. (1978) The Aging Reproductive System. Raven Press, New York.

Schwartz, D., Mayaux, M.-J., Spira, A., Moscato, M.-L., Jouannet, P., Czyglik, F. and David, G. (1983) Semen characteristics as a function of age in 833 fertile men. Fertil. Steril., 39, 530–535.[ISI][Medline]

Singer, R., Sagiv, M., Levinsky, H. and Allalouf, D. (1990) Andrological parameters in men with high sperm counts and possible correlation with age. Arch. Androl., 24, 107–111.[ISI][Medline]

Spandorfer, S.D., Avrech, O.M., Colombero, L.T., Palermo, G.D. and Rosenwaks, Z. (1998) Effect of parental age on fertilization and pregnancy characteristics in couples treated by intracytoplasmic sperm injection. Hum. Reprod., 13, 334–338.[CrossRef][Medline]

Stanwell-Smith, R. and Hendry, W. (1984) The prognosis of male subfertility: a survey of 1025 men referred to a fertility clinic. Br. J. Urol., 56, 422–428.[ISI][Medline]

Stata Statistical Software: Release 7.0 (1999). College Station, TX, Stata Press.

Templeton, A. (1995) Infertility-epidemiology, aetiology and effective management. Health Bull. (Edinb)., 53, 294–298.

Tietze, C. (1957) Reproductive span and rate of conception among Hutterite women. Fertil. Steril., 8, 89–97.[ISI]

van der Westerlaken, L., Naaktgeboren, N. and Helmerhorst, F. (1998) Evaluation of pregnancy rates after intrauterine insemination according to indication, age, and sperm parameters. J. Assist. Reprod. Genet., 15, 359–364.[CrossRef][ISI][Medline]

Ventura, S., Martin, J., Curtin, S. and Mathews, T. (1997). Report of Final Natality Statistics, 1995. Monthly Vital Statistics Report. Vol. 45, no. 11, suppl. National Center for Health Statistics. Hyattsville, MD.

Wang, C., Chan, S., Leung, A., Ng, R., Ng, M., Tang, L., Ma, H., Tsoi, W. and Kwan, M. (1985) Cross-sectional study of semen parameters in a large group of normal Chinese men. Int. J. Androl., 8, 257–274.[ISI][Medline]

World Health Organization (1992) Laboratory Manual for the Examination of Human Semen and Sperm-Cervical Mucus Interaction. World Health Organization, Cambridge, UK.

Wyrobek, A.J. (1983) Methods for evaluating the effects of environmental chemicals on human sperm production. Environ. Health Perspect., 48, 53–59.[ISI][Medline]

Wyrobek, A.J. (1993) Methods and concepts in detecting abnormal reproductive outcomes of paternal origin. Reprod. Toxicol., 7, 3–16.[CrossRef]

Wyrobek, A.J., Gordon, L.A., Burkhart, J.G., Francis, M.W., Kapp, R.W., Jr, Letz, G., Malling, H.V., Topham, J.C. and Whorton, M.D. (1983) An evaluation of human sperm as indicators of chemically induced alterations of spermatogenic function. A report of the U.S. Environmental Protection Agency Gene-Tox Program. Mutat. Res., 115, 73–148.[CrossRef][ISI][Medline]

Submitted on September 6, 2002; accepted on November 17, 2002.