Advertisement

Age thresholds for changes in semen parameters in men

      Objective

      To determine whether age thresholds for elements of semen quality exist.

      Design

      Retrospective analysis (covariance and point-change analysis) of results of 4,822 semen analyses and 259 fluorescence in situ hybridization (FISH) analyses.

      Setting

      Reference laboratory within an infertility clinic.

      Patient(s)

      A total of 5,081 men aged 16.5–72.3 years.

      Intervention(s)

      None.

      Main Outcome Measure(s)

      Ejaculate volume, sperm concentration, sperm motility, sperm motion parameters, strict morphology, and results of FISH analysis.

      Result(s)

      Measured parameters of ejaculates did not change before 34 years of age. Immediately thereafter, total sperm numbers (and total motile) declined. Sperm concentration and the proportion of sperm of normal morphology declined after 40 years. Sperm motility and progressive parameters of motile sperm fell after 43 years and ejaculate volume after 45 years. The ratio of Y:X-bearing sperm in ejaculates decreased only after 55 years.

      Conclusion(s)

      Our findings project a declining likelihood of pregnancy following intercourse with men >34 years old, independent from the woman’s age and increasing with advancing age. Age-related mechanisms associated with this oligoasthenoteratozoospermic progression are discussed.

      Key Words

      Discuss: You can discuss this article with its authors and with other ASRM members at http://fertstertforum.com/stoneba-aging-semen-morphology/
      The ongoing worldwide rise in birth rates to older fathers (
      • Humm K.C.
      • Sakkas D.
      Role of increased male age in IVF and egg donation: is sperm DNA fragmentation responsible?.
      ,
      • Kuhnert B.
      • Nieschlag E.
      Reproductive functions of the ageing male.
      ) is of concern because of evidence that advanced paternal age might (independently from maternal age) be associated with an increased risk of pregnancy loss (
      • de La Rochebrochard E.
      • Thonneau P.
      Paternal age and maternal age are risk factors for miscarriage; results of a multicentre European study.
      ) and a broad range of developmental, morphologic, and neurologic disorders of the newborn (
      • Lian Z.H.
      • Zack M.M.
      • Erickson J.D.
      Paternal age and the occurrence of birth defects.
      ). Age-related processes in the male that could raise the risk of adverse pregnancy outcomes include accumulation of environmental toxins and abuse of alcohol and tobacco (
      • Kidd S.A.
      • Eskanazi B.
      • Wryobek A.J.
      Effects of male age on semen quality and fertility: a review of the literature.
      ), which have now been associated with exponentially rising frequencies of de novo mutations in offspring of older fathers (
      • Kong A.
      • Frigge M.L.
      • Masson G.
      • Besenbacher S.
      • Sulem P.
      • Magnusson G.
      • et al.
      Rate of de novo mutations, father’s age, and disease risk.
      ). In this respect, spermatogenesis remains a sensitive biomarker of toxicity of the broad range of hazardous/mutagenic environmental and industrial agents able to cross the blood-testis barrier (
      • Sharpe R.M.
      Environmental/lifestyle effects on spermatogenesis.
      ). Patterns of changes in spermatogenesis may therefore parallel age-related contraction of (
      • Neaves W.B.
      • Johnson L.
      • Porter J.C.
      • Parker Jr., C.R.
      • Petty S.
      Leydig cell numbers, daily sperm production, and serum gonadotropin levels in aging men.
      ), and other less evident changes in, the germinal epithelium (
      • Mladenovic I.
      • Micic S.
      • Papic N.
      • Genbacev O.
      • Marinkovic B.
      Sperm morphology and motility in different age populations.
      ). In this respect, many have reported lower ejaculate volumes (
      • Kidd S.A.
      • Eskanazi B.
      • Wryobek A.J.
      Effects of male age on semen quality and fertility: a review of the literature.
      ), lower sperm concentrations and total sperm numbers (
      • Pasqualotto F.F.
      • Sobreiri B.P.
      • Hallak J.
      • Pasqualotto E.B.
      • Lucon A.M.
      Sperm concentration and normal sperm morphology decrease and follicle-stimulating hormone level increases with age.
      ), lower sperm motility (
      • Sloter E.
      • Schmidt T.E.
      • Marchetti F.
      • Eskenazi B.
      • Nath J.
      • Wryobek A.J.
      Quantitative effects of male age on sperm motion.
      ), and lower proportions of sperm of normal morphology (
      • Kidd S.A.
      • Eskanazi B.
      • Wryobek A.J.
      Effects of male age on semen quality and fertility: a review of the literature.
      ) in older men. Pasqualotto et al. (
      • Pasqualotto F.F.
      • Sobreiri B.P.
      • Hallak J.
      • Pasqualotto E.B.
      • Lucon A.M.
      Sperm concentration and normal sperm morphology decrease and follicle-stimulating hormone level increases with age.
      ), using data from a relatively small group of patients presenting for vasectomy, identified age thresholds of >40 years for ejaculate volume and >45 years for sperm concentration and motility. Other studies that have established an age effect on any semen parameter have assumed, or inferred, a linear trend across broad age ranges (
      • Levitas E.
      • Lunenfeld E.
      • Weisz N.
      • Friger M.
      • Potashnik G.
      Relationship between age and semen parameters in men with normal sperm concentration: analysis of 6022 semen samples.
      ) with no evidence for age thresholds (
      • Sloter E.
      • Schmidt T.E.
      • Marchetti F.
      • Eskenazi B.
      • Nath J.
      • Wryobek A.J.
      Quantitative effects of male age on sperm motion.
      ). Some have found semen quality to be independent from age (
      • Kalyani R.
      • Basavaraj P.B.
      • Kumar M.L.
      Factors influencing quality of semen: a two year prospective study.
      ), whereas others have shown independence in (young) donors and dependence in older patients (
      • Gandini L.
      • Lombardo F.
      • Culasso F.
      • Dondero F.
      • Lenzi A.
      Myth and reality of the decline in semen quality: an example of the relativity of data interpretation.
      ). As concluded recently, literature on the impact of paternal age on semen parameters remains inconclusive (
      • Humm K.C.
      • Sakkas D.
      Role of increased male age in IVF and egg donation: is sperm DNA fragmentation responsible?.
      ).
      By examining age-related patterns of change in semen parameters in a large standardized dataset, the present study aimed to forge consensus on age thresholds for elements of semen quality, and to determine whether these thresholds coincide with those reported for pregnancy outcomes.

      Subjects and methods

      This study was a noninterventional retrospective analysis of 4,822 consecutive semen analyses performed from January 2007 to December 2012 in our clinical reference laboratory. During this time, patients were referred from 240 physicians nationally, most referrals being from southern California. Although neither the patients nor their referring physicians were specifically polled on justification for seeking semen analysis, most analyses were performed in conjunction with sperm freezing for third-party cycles (egg donor and/or surrogate) or were related to infertility work-up of the patient and his partner. Subfertility of the male partner would be a factor in approximately one-half of the latter category. The study population would therefore have leaned toward subfertility and was not a random cross-section of all men of reproductive age.
      This study analyzed existing data which was exported without identifiers that could be linked to the subjects. Results of this study will not be submitted to the Food and Drug Administration for marketing approval. The Western Institutional Review Board determined that this retrospective study met the conditions for exemption under 45 CFR §46.101(b)(4).
      Because the number of days abstinence is known to influence many semen characteristics (
      • Jouannet P.
      • Czyglik F.
      • David G.
      Study of a group of 484 men. I. Distribution of semen characteristics.
      ), patients were routinely instructed to abstain from intercourse for 3–5 days before their scheduled analysis. Samples were produced at the laboratory by masturbation, and were allowed to liquefy at 35°C for 20 minutes immediately after collection.

       Semen Analysis

      After liquefaction, sperm concentrations, head morphometry, and motion parameters were determined with the use of a Hamilton Thorne IVOS computerized semen analyzer (CASA; Hamilton Thorne Research) with disposable counting chambers (Cell-Vu). All CASA analyses were performed by the same two technicians using the same instrument. Our laboratory maintained compliance with biennial proficiency challenges for semen analysis (Fertility Solutions and American Association of Bioanalysts) throughout the study period. By analyzing semen samples through stepwise dilution series, we have established the reportable range for this instrument to be 7–86 million sperm/mL. Across this range, precision is consistently >90% with >88% accuracy. Samples with sperm concentrations <7 million/mL were scored by hemacytometer; samples with concentrations >86 million/mL were reanalyzed after step-wise dilution with HEPES-buffered human tubal fluid (HTF) supplemented with 5% (v/v) serum substitute supplement (Irvine Scientific). Calibration of the CASA and verification of its reportable range were verified on each day of use with latex beads (Accu-Beads; Hamilton Thorne). Sperm were classified as motile only when their path velocity exceeded 5 μm/s. Those sperm with path velocities >25 μm/s were classified as “rapidly motile” and those faster than 25 μm/s with >80% linearity (straight line velocity/path velocity >0.8) were classified as “progressively motile.” Sperm viability was determined by microscopic scoring of sperm immediately after admixture of 3 μL liquefied semen with 3 μL 0.5% (v/v in saline) eosin B (Sigma). Immature germ cells were identified after staining with peroxide/ortho-toluidine blue (
      • Nahoum C.R.D.
      • Cardozo D.
      Staining for volumetric count of leucocytes in semen and prostate-vesicular fluid.
      ). Sperm morphology was determined by strict criteria (
      • Menkveld R.H.
      • Stander F.S.H.
      • Kotze T.J.
      • Kruger T.F.
      • Van Zyl J.A.
      The evaluation of morphological characteristics of human spermatozoa according to stricter criteria.
      ) after staining of fixed semen smears with Papanicolaou's stain (Spermac; Fertipro).

       Fluorescent in Situ Hybridization Analysis

      FISH analyses were performed on ejaculates from an additional 259 men presenting for routine screening before in vitro fertilization. Liquefied semen was initially washed 3 times with 0.1 mol/L phosphate buffer supplemented with bovine serum albumin (1 mg/mL), then treated with 6 mmol/L EDTA at 37°C. After centrifugation, cells were fixed in suspension (Carnoy) and spread onto positively charged slides, air dried, then aged overnight at room temperature and for >1 hour at 37°C in sodium chloride/sodium citrate (SSC) buffer. Sperm heads were decondensed for 5 minutes in 25 mmol/L DL-dithiothreitol (Sigma) in 1% Triton-100 (Fisher Scientific) at 37°C, then incubated for a further 1 hour in 10% (w/v) pepsin (Sigma) in 0.1 N HCl. Nuclear DNA was denatured in 70% (v/v in SSC) formamide (Sigma) at 73°C and dehydrated through an ethanol series at 4°C immediately preceding overnight hybridization with chromosome enumeration probe (18/X/Y) and locus specific indicator (13/21) probes (Vysis) at 37°C. Slides were finally washed in 50% (v/v in SSC) formamide, NP-40, and SSC at 46°C, counterstained with 125 ng/mL 4,6-diamidino-2-phenylindole in phenylenediamine dihydrochloride and glycerol, then enumerated under fluorescence microscopy. Scoring of fluorescent signals required separation by a minimal distance of one signal domain. All FISH analyses were conducted by the same technician (B.A.S.).

       Statistics

      Data, without transformation, were tested for homogeneity of variance with the use of a modified Bartlett test (
      • Levene H.
      Robust tests for equality of variances.
      ), in which an analysis of variance (ANOVA) is performed on the absolute differences between the dependent variable values and their respective cell means. In such testing, a variable is derived to represent the absolute difference for each observation, the one-way comparison of these values between all cells being the test of homogeneity of variance. This method for testing assumptions of variance is more powerful and more robust against nonnormality than conventional Bartlett or Cochran tests (
      • Levene H.
      Robust tests for equality of variances.
      ). All raw and derived data that met assumptions of independence, normality, and homogeneity of variance (dependent variables in Table 1) were then analyzed by ANOVA with the use of a commercial software package (Crunch Software). The number of days of abstinence was entered as a covariate in the analysis (
      • Jouannet P.
      • Czyglik F.
      • David G.
      Study of a group of 484 men. I. Distribution of semen characteristics.
      ). For the initial ANOVA, records were first ranked by age, then age brackets were set (Table 1) to create eight consecutive age subgroups, each containing equivalent numbers of observations (602 or 603 per group). When F ratios were significant, between-group differences were compared by REGWF, a step-down multiple-stage F test in which P values for individual comparisons are never smaller than the P value for the overall test (
      • Welch R.E.
      Stepwise multiple comparison procedures.
      ).
      Table 1Summary of analysis of covariance (days of abstinence as covariate).
      Age bracket (y; 602–603 records per bracket)P value of F ratio
      16.5–31.231.3–33.833.9–35.835.9–37.938.0–40.040.1–42.542.6–46.546.6–72.3
      Days of abstinence (covariate)3.95 ± 0.113.97 ± 0.093.93 ± 0.083.91 ± 0.074.13 ± 0.234.05 ± 0.104.26 ± 0.174.16 ± 0.14.506
      Ejaculate volume (mL)3.45 ± 0.073.44 ± 0.063.35 ± 0.063.30 ± 0.073.29 ± 0.073.22 ± 0.142.92 ± 0.06
      Age brackets in which average values consistently and significantly declined (REGWF).
      2.49 ± 0.07
      Age brackets in which average values consistently and significantly declined (REGWF).
      <.0001
      CASA
       Concentration (millions/mL)61.5 ± 1.764.2 ± 1.664.1 ± 1.865.5 ± 1.761.1 ± 1.662.1 ± 1.762.4 ± 1.756.9 ± 1.6
      Age brackets in which average values consistently and significantly declined (REGWF).
      .003
       Total sperm (millions)204 ± 7213 ± 7211 ± 8205 ± 6187 ± 6
      Age brackets in which average values consistently and significantly declined (REGWF).
      183 ± 6
      Age brackets in which average values consistently and significantly declined (REGWF).
      176 ± 6
      Age brackets in which average values consistently and significantly declined (REGWF).
      136 ± 5
      Age brackets in which average values consistently and significantly declined (REGWF).
      <.0001
       Sperm motility (%)61.4 ± 1.061.7 ± 1.063.2 ± 1.062.2 ± 1.059.0 ± 1.158.1 ± 1.157.3 ± 1.148.1 ± 1.2
      Age brackets in which average values consistently and significantly declined (REGWF).
      <.0001
       Total motile (millions)146 ± 6155 ± 6153 ± 7148 ± 6131 ± 5124 ± 5
      Age brackets in which average values consistently and significantly declined (REGWF).
      119 ± 5
      Age brackets in which average values consistently and significantly declined (REGWF).
      83 ± 4
      Age brackets in which average values consistently and significantly declined (REGWF).
      <.0001
       Rapid motility (%)45.3 ± 0.945.0 ± 0.946.9 ± 0.945.6 ± 0.943.5 ± 0.943.0 ± 1.341.6 ± 0.934.9 ± 0.9
      Age brackets in which average values consistently and significantly declined (REGWF).
      <.0001
       Progressive motility (%)33.2 ± 0.632.6 ± 0.634.1 ± 0.633.2 ± 0.631.8 ± 0.730.9 ± 0.631.1 ± 0.726.1 ± 0.7
      Age brackets in which average values consistently and significantly declined (REGWF).
      <.0001
       Progressive velocity (μm/s)36.7 ± 0.436.2 ± 0.337.1 ± 0.336.8 ± 0.436.2 ± 0.435.7 ± 0.336.2 ± 0.433.9 ± 0.4
      Age brackets in which average values consistently and significantly declined (REGWF).
      <.0001
       Head width/length (%)63.8 ± 0.962.6 ± 0.861.2 ± 0.762.8 ± 0.262.4 ± 0.661.9 ± 0.559.4 ± 0.6
      Age brackets in which average values consistently and significantly declined (REGWF).
      55.8 ± 1.3
      Age brackets in which average values consistently and significantly declined (REGWF).
      <.0001
      Staining and manual microscopy
       Viability (%)59.9 ± 0.658.6 ± 0.658.3 ± 0.659.0 ± 0.657.7 ± 0.756.5 ± 0.653.9 ± 0.6
      Age brackets in which average values consistently and significantly declined (REGWF).
      48.3 ± 0.8
      Age brackets in which average values consistently and significantly declined (REGWF).
      <.0001
       Normal morphology (%)13.1 ± 0.413.4 ± 0.413.8 ± 0.413.4 ± 0.412.9 ± 0.413.4 ± 0.413.8 ± 0.411.9 ± 0.3
      Age brackets in which average values consistently and significantly declined (REGWF).
      .0191
      FISH
       Aneuploidy (X/Y/13/18/21)2.00 ± 0.192.94 ± 0.442.70 ± 0.452.43 ± 0.422.34 ± 0.242.85 ± 0.452.62 ± 0.282.36 ± 0.29NS
       Diploid (%)0.14 ± 0.140.81 ± 0.350.29 ± 0.160.38 ± 0.150.16 ± 0.040.54 ± 0.280.18 ± 0.050.12 ± 0.04NS
       Y:X ratio1.06 ± 0.071.08 ± 0.151.08 ± 0.161.05 ± 0.081.06 ± 0.071.03 ± 0.131.04 ± 0.091.00 ± 0.07
      Age brackets in which average values consistently and significantly declined (REGWF).
      .0409
      Note: CASA = computer-assisted semen analysis; FISH = fluorescence in situ hybridization.
      Age brackets in which average values consistently and significantly declined (REGWF).
      The data were also analyzed by least-squares linear regression. Those variables found to significantly correlate with age across the entire age range were then further analyzed by change-point regression analysis. Although developed primarily for economic models (
      • Krishnaiah P.R.
      • Miao B.Q.
      Review about estimation of change points.
      ), change-point regression challenges the general hypothesis that a regression slope might change at a given abscissal (age) threshold, and need not be constant. In the present study, change-point regression analysis complemented the ANOVA by enabling definition of discrete age thresholds not evident through the ANOVA or through simple least-squares linear regression. That is, change-point analysis is relevant to this study’s pursuit of age thresholds at which the overall correlation between age and a semen parameter might change (steepen). We used a change-point regression algorithm previously adapted for physiology (
      • Julios S.A.
      Inference and estimation in a changepoint regression problem.
      ), applying thresholds estimated from the ANOVA as initial parameter estimates.

      Results

      Results of the initial ANOVA are detailed in Table 1. Asterisked values in Table 1 indicate the age brackets in which average values for the measured parameters consistently and significantly declined (REGWF). Consistent with the ANOVA, linear regression analysis across all ages revealed that the number of days of abstinence increased on average by only 0.01 for each additional year of age, the regression correlation being insignificant (P=.15). This likely reflects compliance of patients with instructions provided by the laboratory at the time of scheduling (3–5 days abstinence) rather than otherwise impulsive periods of abstinence, which have been found to increase with advancing age (
      • Pasqualotto F.F.
      • Sobreiri B.P.
      • Hallak J.
      • Pasqualotto E.B.
      • Lucon A.M.
      Sperm concentration and normal sperm morphology decrease and follicle-stimulating hormone level increases with age.
      ,
      • Sloter E.
      • Schmidt T.E.
      • Marchetti F.
      • Eskenazi B.
      • Nath J.
      • Wryobek A.J.
      Quantitative effects of male age on sperm motion.
      ,
      • Jouannet P.
      • Czyglik F.
      • David G.
      Study of a group of 484 men. I. Distribution of semen characteristics.
      ). To satisfy statistical doubt on the impact of abstinence on findings, the number of days of abstinence was retained as a covariate in the ANOVA but did not change conclusions from post hoc testing.
      By both ANOVA (Table 1) and change-point analysis (Table 2), all measured parameters were uniform through age 34 years. Although statistically significant depression in sperm concentration and ejaculate volume did not appear until 40 and 45 years, respectively (Tables 1 and 2), calculated total numbers of sperm began to decline at 34 years. That is, total sperm output was more statistically sensitive to aging than either of its mathematical determinants. All sperm motility parameters (motility, rapid motility, progressive motility, and average progressive velocity) began to decline at 43 years. Coincident with declining total sperm numbers, total numbers of motile sperm (motile and progressively motile) were also lower after 34 years. Sperm head membrane viability (eosin exclusion) diminished in patients only after 45 years.
      Table 2Summary of change-point regression analysis, performed only on those dependent variables with which there was an overall significant correlation with patient age.
      Dependent variableLocation of the change-point (y)Slope of the regression line beyond the change-pointSignificance (P) of the correlationProjected change per year beyond the change-point
      Calculated from the slope of the regression beyond the change-point and the average value for each dependent variable at each respective change-point.
      Total sperm (millions)34−3.604<.001−1.71%
      Total motile (millions)34−3.544<.001−2.30%
      Total progressive (millions)34−1.832<.001−2.61%
      Sperm concentration (millions/mL)40−0.483<.001−0.78%
      Normal morphology (%)40−0.111.002−0.84%
      Head width/length (%)41−0.509<.001−0.83%
      Sperm motility (%)43−0.990<.001−1.74%
      Rapid motility (%)43−0.779<.001−1.90%
      Progressive motility (%)43−0.585<.001−1.95%
      Progressive velocity (μm/second)43−0.282<.001−0.81%
      Ejaculate volume (mL)45−0.043<.001−1.48%
      Membrane viability (%)45−0.770<.001−1.45%
      Ratio of Y- to X-bearing sperm55−0.054<.05−5.05%
      Note: Dependent variables are listed in ascending order of the derived age change-point.
      a Calculated from the slope of the regression beyond the change-point and the average value for each dependent variable at each respective change-point.
      The fall in sperm head:width ratio to <62% in sperm of ejaculates from patients >41 years old indicates abnormal head elongation, the onset of which closely coincided with lower proportions of sperm of normal morphology by strict criteria (at >40 years; Tables 1 and 2).
      Proportions of aneuploid or diploid sperm did not change significantly over the age range studied. The proportion of Y- to X-bearing sperm was lower in the oldest patient subgroup (>46.5 years), with change-point analysis establishing a break point at 55 years (Table 2). As incidental observations during scoring, nuclei of sperm with large heads (>7 μm in length or width) were, without exception, diploid; nuclei of sperm with round heads were typically, but not exclusively, aneuploid.

      Discussion

      Earlier attempts to draw conclusions from pooled data from different publications (
      • Kidd S.A.
      • Eskanazi B.
      • Wryobek A.J.
      Effects of male age on semen quality and fertility: a review of the literature.
      ,
      • Sloter E.
      • Schmidt T.E.
      • Marchetti F.
      • Eskenazi B.
      • Nath J.
      • Wryobek A.J.
      Quantitative effects of male age on sperm motion.
      ) or from the same investigators at different locations (
      • Fisch H.
      • Goluboff E.T.
      • Olson J.H.
      • Feldshuh J.
      • Broder S.J.
      • Barad D.H.
      Semen analyses in 1,283 men from the United States over a 25-year period: no decline in quality.
      ) have been plagued by inconsistencies in methods, personnel, and instrumentation. The lack of standardization in analytic techniques weakened statistical analysis of pooled data by compounding the existent inherent variability in measured semen characteristics. In the present analysis, methods, personnel, and instrumentation were consistent throughout the study period.
      With the exception of lower average measured values for ejaculates from patients older than 46.5 years, average values for volume, sperm concentration, motility, and total sperm numbers in all age groups in the present study (Table 1) were within the World Health Organization’s respective 25th and 75th percentile ranges for unscreened men from the general population (
      • Cooper T.G.
      • Noonan E.
      • von Eckardstein S.
      • Auger J.
      • Gordon Baker H.W.
      • Behre H.M.
      • et al.
      World Health Organization reference values for human semen characteristics.
      ). That is, although the proportion of subfertile subjects in our study population should be higher than that in the general population, this weighting did not result in a substantial shift in determined average values. It is not known to what extent, if any, age change-points might have been affected by a weighting of our population toward subfertility.
      Our analysis revealed that consistent significant depression in numbers of sperm (total and total motile) appeared after ∼34 years (Tables 1 and 2). This threshold closely parallels earlier projections (
      • Levitas E.
      • Lunenfeld E.
      • Weisz N.
      • Friger M.
      • Potashnik G.
      Relationship between age and semen parameters in men with normal sperm concentration: analysis of 6022 semen samples.
      ), including reported maximal total sperm numbers at 31–35 years in patients presenting for vasectomy (
      • Pasqualotto F.F.
      • Sobreiri B.P.
      • Hallak J.
      • Pasqualotto E.B.
      • Lucon A.M.
      Sperm concentration and normal sperm morphology decrease and follicle-stimulating hormone level increases with age.
      ). Studies citing later declines (at ∼40 years) acknowledged residual confounding factors in their analyses (
      • Gandini L.
      • Lombardo F.
      • Culasso F.
      • Dondero F.
      • Lenzi A.
      Myth and reality of the decline in semen quality: an example of the relativity of data interpretation.
      ). Motility and progressive characteristics of motile sperm were affected at 43 years of age (Table 2), coincident with rising levels of reactive oxygen species (ROS) in semen in this same age range (
      • Coccuza M.
      • Athayde K.S.
      • Agarwal A.
      • Sharma R.
      • Pagani R.
      • Lucon A.M.
      • et al.
      Age- related increase in reactive oxygen species in neat semen in healthy fertile men.
      ). Through establishing negative correlations between seminal ROS levels and sperm concentration and motility, numerous investigators (
      • Coccuza M.
      • Athayde K.S.
      • Agarwal A.
      • Sharma R.
      • Pagani R.
      • Lucon A.M.
      • et al.
      Age- related increase in reactive oxygen species in neat semen in healthy fertile men.
      ) have projected that rising seminal ROS levels may be a possible causative factor for incremental asthenozoospermia with aging. Coincident with declining sperm production (Table 2), serum FSH levels in healthy men also have been reported to rise more steeply after ∼40 years (
      • Pasqualotto F.F.
      • Sobreiri B.P.
      • Hallak J.
      • Pasqualotto E.B.
      • Lucon A.M.
      Sperm concentration and normal sperm morphology decrease and follicle-stimulating hormone level increases with age.
      ), reflecting progressive refractoriness of the contracting germinal epithelium (
      • Neaves W.B.
      • Johnson L.
      • Porter J.C.
      • Parker Jr., C.R.
      • Petty S.
      Leydig cell numbers, daily sperm production, and serum gonadotropin levels in aging men.
      ) to gonadotropic support.
      Declining sperm numbers and motility can relate to the likelihood of conception following unprotected intercourse or intrauterine insemination (IUI) (
      • Stone B.A.
      • Vargyas J.M.
      • Ringler G.E.
      • Stein A.L.
      • Marrs R.P.
      Determinants of the outcome of intrauterine insemination (IUI); analysis of outcomes of 9,963 consecutive cycles.
      ), the threshold numbers of motile sperm below which pregnancy rates decline being ∼40 million for intercourse (
      • Cooper T.G.
      • Noonan E.
      • von Eckardstein S.
      • Auger J.
      • Gordon Baker H.W.
      • Behre H.M.
      • et al.
      World Health Organization reference values for human semen characteristics.
      ) or ∼4 million for IUI (
      • Stone B.A.
      • Vargyas J.M.
      • Ringler G.E.
      • Stein A.L.
      • Marrs R.P.
      Determinants of the outcome of intrauterine insemination (IUI); analysis of outcomes of 9,963 consecutive cycles.
      ). Numbers of total motile sperm were <40 million in 48% of ejaculates from our oldest age subgroup, having increased from 25% in the youngest subgroup. That is, although average numbers of motile sperm in ejaculates from patients >46.5 years exceeded 40 million (Table 1), the proportion of patients with compromising total motile sperm numbers increased with advancing age. There are little other published data against which to project the impact of progressive age-related oligoasthenozoospermia on pregnancy outcomes. Most population studies that determined poorer pregnancy outcomes from older men, some of which have determined 40 years as a threshold (
      • Humm K.C.
      • Sakkas D.
      Role of increased male age in IVF and egg donation: is sperm DNA fragmentation responsible?.
      ), did not assess outcomes from continuous age spectra (
      • Ford W.C.
      • North K.
      • Taylor H.
      • Farrow A.
      • Hull M.G.
      • Golding J.
      Increasing paternal age is associated with delayed conception in a large population of fertile couples: evidence for declining fecundity in older men.
      ). Rather, such studies have reported outcomes surrounding a single selected age (
      • Frattarelli J.L.
      • Miller K.A.
      • Miller B.T.
      • Elkind-Hirsch K.
      • Scott R.T.
      Male age negatively impacts embryo development and reproductive outcome in donor oocyte assisted reproductive technology cycles.
      ) or have compared outcomes from old with those of chronologically remote young patient subgroups (
      • Vagnini L.
      • Baruffi R.L.
      • Mauri A.L.
      • Petersen C.G.
      • Massaro F.C.
      • Pontes A.
      • et al.
      The effects of male age on sperm DNA damage in an infertile population.
      ). An exception is Klonoff-Cohen et al.’s analysis (
      • Klonoff-Cohen H.S.
      • Natarajan L.
      The effect of advancing paternal age on pregnancy and live birth rates in couples undergoing in vitro fertilization or gamete intrafallopian transfer.
      ) of outcomes from three contiguous paternal age groups, in which birth rates relative to paternal age were declining before 40 years. Because a selected subset of motile sperm are used for fertilization in IVF/intracytoplasmic sperm injection (ICSI) cycles, this and other reports of poor embryo quality and lower pregnancy rates following assisted reproduction technologies (particularly those with donor oocytes [
      • Frattarelli J.L.
      • Miller K.A.
      • Miller B.T.
      • Elkind-Hirsch K.
      • Scott R.T.
      Male age negatively impacts embryo development and reproductive outcome in donor oocyte assisted reproductive technology cycles.
      ]) with processed semen from older men provide substantive evidence that those functional properties of selected sperm which influence embryo quality (e.g., chromosomal/DNA integrity) are compromised with ageing (
      • Vagnini L.
      • Baruffi R.L.
      • Mauri A.L.
      • Petersen C.G.
      • Massaro F.C.
      • Pontes A.
      • et al.
      The effects of male age on sperm DNA damage in an infertile population.
      ). Declining numbers of total and of total motile sperm in ejaculates of men aged >34 years (Table 2) would compound the impact of these changes on pregnancy outcomes following intercourse or IUI with sperm providers <40 years old and are thereby consistent with Klonoff-Cohen et al.’s findings (
      • Klonoff-Cohen H.S.
      • Natarajan L.
      The effect of advancing paternal age on pregnancy and live birth rates in couples undergoing in vitro fertilization or gamete intrafallopian transfer.
      ). Other studies have found no paternal age effect on pregnancy rates following IVF/ICSI, including those of donor oocyte cycles (
      • Whitcomb B.W.
      • Turzanski-Fortner R.
      • Richter K.S.
      • Kipersztok S.
      • Stillman R.J.
      • Levy M.J.
      • et al.
      Contribution of male age to outcomes in assisted reproductive technologies.
      ).
      Sperm head elongation and the overall proportion of sperm of normal morphology did not fall significantly until after 40 years (Tables 1 and 2). Earlier analyses of teratozoospermia against advancing age (
      • Pasqualotto F.F.
      • Sobreiri B.P.
      • Hallak J.
      • Pasqualotto E.B.
      • Lucon A.M.
      Sperm concentration and normal sperm morphology decrease and follicle-stimulating hormone level increases with age.
      ,
      • Auger J.
      • Kuntsmann J.M.
      • Czyglik F.
      • Jouannet P.
      Decline in semen quality among fertile men in Paris during the past 20 years.
      ) established a threshold at ∼45 years. Annual rates of decline reported earlier (∼0.9% [
      • Klonoff-Cohen H.S.
      • Natarajan L.
      The effect of advancing paternal age on pregnancy and live birth rates in couples undergoing in vitro fertilization or gamete intrafallopian transfer.
      ]) closely paralleled our determination of 0.8%/year (Table 2). Through established association with teratozoospermia, rising rates of chromosomal abnormality (fragmentation, terminal deoxynucleotide transferase–mediated dUTP nick-end labeling) and aneuploidy (
      • Martin R.H.
      • Spriggs E.
      • Ko E.
      • Radenmaker A.W.
      The relationship between paternal age, sex ratios, and aneuploidy frequencies in human sperm, as assessed by multicolor FISH.
      ) should be expected in sperm of age-advanced males (
      • Martin R.H.
      • Rademaker A.W.
      The effect of age on the frequency of sperm chromosomal abnormalities in normal men.
      ). Our analysis of chromosomes 13, 18, 21, X, and Y (Table 1) established an overall incidence of aneuploidy for this chromosome subset to be ∼3%, consistent with earlier published ranges for all chromosomes of 2%–13% in normal fertile men (
      • Pang M.G.
      • Hoegerman S.F.
      • Cuticchia A.J.
      • Moon S.Y.
      • Doncel G.F.
      • Acosta A.A.
      • et al.
      Detection of aneuploidy for chromosomes 4, 6, 7, 8, 9,10,11,12,13,17,18, 21, X and Y by fluorescence in-situ hybridization in spermatozoa from nine patients with oligoasthenoteratozoospermia undergoing intracytoplasmic sperm injection.
      ). The frequency of aneuploidy, higher in gonosomes than in autosomes (
      • Pang M.G.
      • Hoegerman S.F.
      • Cuticchia A.J.
      • Moon S.Y.
      • Doncel G.F.
      • Acosta A.A.
      • et al.
      Detection of aneuploidy for chromosomes 4, 6, 7, 8, 9,10,11,12,13,17,18, 21, X and Y by fluorescence in-situ hybridization in spermatozoa from nine patients with oligoasthenoteratozoospermia undergoing intracytoplasmic sperm injection.
      ), was independent from patient age (Table 2) despite the higher incidence of teratozoospermia in the older patients (Table 1). This finding aligns with those of others (
      • Martin R.H.
      • Spriggs E.
      • Ko E.
      • Radenmaker A.W.
      The relationship between paternal age, sex ratios, and aneuploidy frequencies in human sperm, as assessed by multicolor FISH.
      ), but is contrary to claimed positive correlations between sperm aneuploidy and age (
      • Martin R.H.
      • Rademaker A.W.
      The effect of age on the frequency of sperm chromosomal abnormalities in normal men.
      ). Our incidental observation that nuclei of all sperm with large heads (>7 μmol/L) were diploid, regardless of patient age, is of importance to the selection of sperm for ICSI.
      In an early analysis, Novitski (
      • Novitski E.
      The dependence of the secondary sex ratio in humans on the age of the father.
      ) demonstrated a declining secondary sex ratio that correlated more closely with advancing paternal age than with advancing maternal age. Novitski and Sandler (
      • Novitski E.
      • Sandler L.
      The relationship between paternal age, birth order and the secondary sex ratio in humans.
      ) then postulated that the changing sex ratio might reflect a change in the relative frequency of Y- and X-bearing sperm. To our knowledge, our data (Tables 1 and 2) are the first to establish a negative correlation between the ratio of Y:X-bearing sperm and age (P=.03; Table 2), the ratio of Y:X-bearing sperm in ejaculates from older men (1.0; Table 1) closely paralleling the secondary sex ratio for children of older fathers (
      • Ruder A.
      Paternal age and birth order effect on the human secondary sex ratio.
      ). The ratio of Y:X-bearing sperm in ejaculates from men <46 years of age was consistently ∼1.06 (Table 1), as previously reported (
      • Martin R.H.
      • Spriggs E.
      • Ko E.
      • Radenmaker A.W.
      The relationship between paternal age, sex ratios, and aneuploidy frequencies in human sperm, as assessed by multicolor FISH.
      ).
      Our analysis of total sperm numbers in ejaculates indicates that daily sperm production in men declines after ∼34 years of age (total sperm numbers), falling by ∼2% per year thereafter (Table 2). Sperm concentration and morphology are affected about 6 years later, each diminishing by ∼0.8% per year. The proportion of motile sperm and the progressive parameters of motile sperm decline by ∼2% and 0.8%, respectively, per year, but only after 43 years (Table 2). Recognizing compounding functional changes that occur with human sperm through this same age range (
      • Klonoff-Cohen H.S.
      • Natarajan L.
      The effect of advancing paternal age on pregnancy and live birth rates in couples undergoing in vitro fertilization or gamete intrafallopian transfer.
      ), our findings project a declining likelihood of pregnancy following intercourse with men older than ∼34 years. This projection is aligned with earlier determinations that pregnancy outcomes, adjusted for maternal age, decreased when paternal age exceeded 35 years (
      • Dunson D.B.
      • Colombo B.
      • Baird D.D.
      Changes with age in the level and duration of fertility in the menstrual cycle.
      ,
      • Mathieu C.
      • Ecochard R.
      • Bied V.
      • Lornage J.
      • Czyba J.C.
      Cumulative conception rate following intrauterine artificial insemination with husband’s spermatozoa: influence of husband’s age.
      ). This finding is also of interest considering the recent report that the incidence of paternally derived de novo point mutations in offspring increases steeply when paternal age exceeds 35 years (
      • O’Roak B.J.
      • Vives L.
      • Girirajan S.
      • Karakoc E.
      • Krumm N.
      • Coe B.P.
      • et al.
      Sporadic autism exomes reveal a highly interconnected protein network of de novo mutations.
      ).
      Age-associated mechanisms behind the progressive oligoasthenoteratozoospermia characterized in our study could include rising seminal ROS (
      • Coccuza M.
      • Athayde K.S.
      • Agarwal A.
      • Sharma R.
      • Pagani R.
      • Lucon A.M.
      • et al.
      Age- related increase in reactive oxygen species in neat semen in healthy fertile men.
      ). Because sperm numbers in ejaculates reflect activity of the testicular germinal epithelium, whereas sperm motility is acquired in the epididymis, our findings also indicate similarly age-related, but independent, groups of mechanisms negatively affecting testicular and epididymal function. In this respect, as has been shown in nonhuman scrotal mammals, spermatogenesis appears to be more sensitive to heating than the acquisition of motility (
      • Stone B.A.
      Thermal characteristics of the testis and epididymis of the boar.
      ,
      • Stone B.A.
      Heat-induced infertility of boars: the inter-relationship between depressed sperm output and fertility and an estimate of the critical air temperature above which sperm output is impaired.
      ). Therefore, the present findings may reflect deterioration in local thermoregulation secondary to established age-related arteriosclerotic and other vascularization anomalies in the testis and epididymis. These anomalies could include intimal hyperplasia of spermatic cord vessels (
      • Regadera J.
      • Nistal M.
      • Paniagua R.
      Testis, epididymis, and spermatic cord in elderly men. Correlation of angiographic and histologic studies with systemic arteriosclerosis.
      ) and hyalinization of arterioles (
      • Suoranta H.
      Changes in small blood vessels of the adult human testis in relation to age and to some pathological conditions.
      ) which would affect efficiency of countercurrent heat exchange in the pampiniform plexus (
      • Stone B.A.
      Thermal characteristics of the testis and epididymis of the boar.
      ). Chronologically, these changes parallel established age-related degeneration of the germinal epithelium and atrophy of Leydig cells and other components of the testicular interstitial tissue (
      • Neaves W.B.
      • Johnson L.
      • Porter J.C.
      • Parker Jr., C.R.
      • Petty S.
      Leydig cell numbers, daily sperm production, and serum gonadotropin levels in aging men.
      ), with no coincident change in testis volume (
      • Pasqualotto F.F.
      • Sobreiri B.P.
      • Hallak J.
      • Pasqualotto E.B.
      • Lucon A.M.
      Sperm concentration and normal sperm morphology decrease and follicle-stimulating hormone level increases with age.
      ). Later declines in the ratio of Y:X-bearing sperm and in the proportion of sperm of normal morphology (Tables 1 and 2) (
      • Pasqualotto F.F.
      • Sobreiri B.P.
      • Hallak J.
      • Pasqualotto E.B.
      • Lucon A.M.
      Sperm concentration and normal sperm morphology decrease and follicle-stimulating hormone level increases with age.
      ) would then be consequent to different, and/or additional, age-related phenomena.

      Acknowledgments

      The authors acknowledge support from the many physicians who have entrusted their patients to our laboratories for andrology services.

      References

        • Humm K.C.
        • Sakkas D.
        Role of increased male age in IVF and egg donation: is sperm DNA fragmentation responsible?.
        Fertil Steril. 2013; 99: 30-36
        • Kuhnert B.
        • Nieschlag E.
        Reproductive functions of the ageing male.
        Hum Reprod Update. 2004; 10: 327-329
        • de La Rochebrochard E.
        • Thonneau P.
        Paternal age and maternal age are risk factors for miscarriage; results of a multicentre European study.
        Hum Reprod. 2002; 17: 1649-1656
        • Lian Z.H.
        • Zack M.M.
        • Erickson J.D.
        Paternal age and the occurrence of birth defects.
        Am J Hum Genet. 1986; 39: 648-660
        • Kidd S.A.
        • Eskanazi B.
        • Wryobek A.J.
        Effects of male age on semen quality and fertility: a review of the literature.
        Fertil Steril. 2001; 75: 237-248
        • Kong A.
        • Frigge M.L.
        • Masson G.
        • Besenbacher S.
        • Sulem P.
        • Magnusson G.
        • et al.
        Rate of de novo mutations, father’s age, and disease risk.
        Nature. 2012; 488: 471-475
        • Sharpe R.M.
        Environmental/lifestyle effects on spermatogenesis.
        Philos Trans R Soc Lond B Biol Sci. 2010; 365: 1697-1712
        • Neaves W.B.
        • Johnson L.
        • Porter J.C.
        • Parker Jr., C.R.
        • Petty S.
        Leydig cell numbers, daily sperm production, and serum gonadotropin levels in aging men.
        J Clin Endocrinol Metab. 1984; 59: 756-763
        • Mladenovic I.
        • Micic S.
        • Papic N.
        • Genbacev O.
        • Marinkovic B.
        Sperm morphology and motility in different age populations.
        Arch Androl. 1994; 32: 197-205
        • Pasqualotto F.F.
        • Sobreiri B.P.
        • Hallak J.
        • Pasqualotto E.B.
        • Lucon A.M.
        Sperm concentration and normal sperm morphology decrease and follicle-stimulating hormone level increases with age.
        Brit J Urol Intern. 2005; 96: 1087-1091
        • Sloter E.
        • Schmidt T.E.
        • Marchetti F.
        • Eskenazi B.
        • Nath J.
        • Wryobek A.J.
        Quantitative effects of male age on sperm motion.
        Hum Reprod. 2006; 21: 2868-2875
        • Levitas E.
        • Lunenfeld E.
        • Weisz N.
        • Friger M.
        • Potashnik G.
        Relationship between age and semen parameters in men with normal sperm concentration: analysis of 6022 semen samples.
        Andrologia. 2007; 39: 45-50
        • Kalyani R.
        • Basavaraj P.B.
        • Kumar M.L.
        Factors influencing quality of semen: a two year prospective study.
        Indian J Pathol Microbiol. 2007; 50: 890-895
        • Gandini L.
        • Lombardo F.
        • Culasso F.
        • Dondero F.
        • Lenzi A.
        Myth and reality of the decline in semen quality: an example of the relativity of data interpretation.
        J Endocrinol Invest. 2000; 23: 402-411
        • Jouannet P.
        • Czyglik F.
        • David G.
        Study of a group of 484 men. I. Distribution of semen characteristics.
        Int J Androl. 1981; 4: 440-449
        • Nahoum C.R.D.
        • Cardozo D.
        Staining for volumetric count of leucocytes in semen and prostate-vesicular fluid.
        Fertil Steril. 1980; 34: 68-69
        • Menkveld R.H.
        • Stander F.S.H.
        • Kotze T.J.
        • Kruger T.F.
        • Van Zyl J.A.
        The evaluation of morphological characteristics of human spermatozoa according to stricter criteria.
        Hum Reprod. 1990; 5: 586-592
        • Levene H.
        Robust tests for equality of variances.
        in: Olkin I. Contributions to probability and statistics. Stanford University Press, Palo Alto, California1960: 278-292
        • Welch R.E.
        Stepwise multiple comparison procedures.
        J Am Stat Assoc. 1977; 72: 566-575
        • Krishnaiah P.R.
        • Miao B.Q.
        Review about estimation of change points.
        in: Handbook of statistics. Vol. 7. Elsevier, New York1988: 375-401
        • Julios S.A.
        Inference and estimation in a changepoint regression problem.
        Statistician. 2001; 50: 51-61
        • Fisch H.
        • Goluboff E.T.
        • Olson J.H.
        • Feldshuh J.
        • Broder S.J.
        • Barad D.H.
        Semen analyses in 1,283 men from the United States over a 25-year period: no decline in quality.
        Fertil Steril. 1996; 65: 1009-1014
        • Cooper T.G.
        • Noonan E.
        • von Eckardstein S.
        • Auger J.
        • Gordon Baker H.W.
        • Behre H.M.
        • et al.
        World Health Organization reference values for human semen characteristics.
        Hum Reprod Update. 2010; 16: 231-245
        • Coccuza M.
        • Athayde K.S.
        • Agarwal A.
        • Sharma R.
        • Pagani R.
        • Lucon A.M.
        • et al.
        Age- related increase in reactive oxygen species in neat semen in healthy fertile men.
        Urology. 2008; 71: 490-494
        • Stone B.A.
        • Vargyas J.M.
        • Ringler G.E.
        • Stein A.L.
        • Marrs R.P.
        Determinants of the outcome of intrauterine insemination (IUI); analysis of outcomes of 9,963 consecutive cycles.
        Am J Obstet Gynecol. 1999; 180: 1524-1534
        • Ford W.C.
        • North K.
        • Taylor H.
        • Farrow A.
        • Hull M.G.
        • Golding J.
        Increasing paternal age is associated with delayed conception in a large population of fertile couples: evidence for declining fecundity in older men.
        Hum Reprod. 2000; 15: 1703-1708
        • Frattarelli J.L.
        • Miller K.A.
        • Miller B.T.
        • Elkind-Hirsch K.
        • Scott R.T.
        Male age negatively impacts embryo development and reproductive outcome in donor oocyte assisted reproductive technology cycles.
        Fertil Steril. 2008; 90: 97-103
        • Vagnini L.
        • Baruffi R.L.
        • Mauri A.L.
        • Petersen C.G.
        • Massaro F.C.
        • Pontes A.
        • et al.
        The effects of male age on sperm DNA damage in an infertile population.
        Reprod Biomed Online. 2007; 15: 514-519
        • Klonoff-Cohen H.S.
        • Natarajan L.
        The effect of advancing paternal age on pregnancy and live birth rates in couples undergoing in vitro fertilization or gamete intrafallopian transfer.
        Am J Obstet Gynecol. 2004; 191: 507-514
        • Whitcomb B.W.
        • Turzanski-Fortner R.
        • Richter K.S.
        • Kipersztok S.
        • Stillman R.J.
        • Levy M.J.
        • et al.
        Contribution of male age to outcomes in assisted reproductive technologies.
        Fertil Steril. 2011; 95: 147-151
        • Auger J.
        • Kuntsmann J.M.
        • Czyglik F.
        • Jouannet P.
        Decline in semen quality among fertile men in Paris during the past 20 years.
        N Engl J Med. 1995; 332: 281-285
        • Martin R.H.
        • Spriggs E.
        • Ko E.
        • Radenmaker A.W.
        The relationship between paternal age, sex ratios, and aneuploidy frequencies in human sperm, as assessed by multicolor FISH.
        Am J Hum Genet. 1995; 57: 1395-1399
        • Martin R.H.
        • Rademaker A.W.
        The effect of age on the frequency of sperm chromosomal abnormalities in normal men.
        Am J Hum Genet. 1987; 41: 484-492
        • Pang M.G.
        • Hoegerman S.F.
        • Cuticchia A.J.
        • Moon S.Y.
        • Doncel G.F.
        • Acosta A.A.
        • et al.
        Detection of aneuploidy for chromosomes 4, 6, 7, 8, 9,10,11,12,13,17,18, 21, X and Y by fluorescence in-situ hybridization in spermatozoa from nine patients with oligoasthenoteratozoospermia undergoing intracytoplasmic sperm injection.
        Hum Reprod. 1999; 14: 1266-1273
        • Novitski E.
        The dependence of the secondary sex ratio in humans on the age of the father.
        Science. 1953; 117: 531-533
        • Novitski E.
        • Sandler L.
        The relationship between paternal age, birth order and the secondary sex ratio in humans.
        Am J Hum Genet. 1956; 221: 123-131
        • Ruder A.
        Paternal age and birth order effect on the human secondary sex ratio.
        Am J Hum Genet. 1985; 37: 362-372
        • Dunson D.B.
        • Colombo B.
        • Baird D.D.
        Changes with age in the level and duration of fertility in the menstrual cycle.
        Hum Reprod. 2002; 17: 1399-1403
        • Mathieu C.
        • Ecochard R.
        • Bied V.
        • Lornage J.
        • Czyba J.C.
        Cumulative conception rate following intrauterine artificial insemination with husband’s spermatozoa: influence of husband’s age.
        Hum Reprod. 1995; 10: 1090-1097
        • O’Roak B.J.
        • Vives L.
        • Girirajan S.
        • Karakoc E.
        • Krumm N.
        • Coe B.P.
        • et al.
        Sporadic autism exomes reveal a highly interconnected protein network of de novo mutations.
        Nature. 2012; 485: 246-250
        • Stone B.A.
        Thermal characteristics of the testis and epididymis of the boar.
        J Reprod Fert. 1981; 63: 551-557
        • Stone B.A.
        Heat-induced infertility of boars: the inter-relationship between depressed sperm output and fertility and an estimate of the critical air temperature above which sperm output is impaired.
        Anim Reprod Sci. 1982; 4: 283-299
        • Regadera J.
        • Nistal M.
        • Paniagua R.
        Testis, epididymis, and spermatic cord in elderly men. Correlation of angiographic and histologic studies with systemic arteriosclerosis.
        Arch Path Lab Med. 1985; 109: 663-667
        • Suoranta H.
        Changes in small blood vessels of the adult human testis in relation to age and to some pathological conditions.
        Pathol Anat. 1971; 352: 165-181