Changes in Semen Analysis over Time: A Temporal Trend Analysis of 20 Years of Subfertile Non-Azoospermic Men

Abstract

Purpose

To examine trends of population-level semen quality over a 20-year period.

Materials and Methods

We performed a retrospective review of data from the andrology lab of a high volume tertiary hospital. All men with semen samples between 2000 and 2019 were included and men with azoospermia were excluded. Semen parameters were reported using the World Health Organization (WHO) 4th edition. The primary outcome of interest was changes in semen parameters over time. Generalized least squares (GLS) with restricted cubic splines were used to estimate average-monthly measurements, adjusting for age and abstinence period. Contrasts of the estimated averages based on GLS between the first and last months of collection were calculated.

Results

A total of 8,990 semen samples from subfertile non-azoospermic men were included in our study. Semen volume decreased over time and estimate average at the beginning and end were statistically different (p<0.001). Similarly sperm morphology decreased over time, with a statistically significant difference between estimated averages from start to finish (p<0.001). Semen pH appeared to be increasing over time, but this difference was not significant over time (p=0.060). Sperm concentration and count displayed an increase around 2003 to 2005, but otherwise remained fairly constant over time (p=0.100 and p=0.054, respectively). Sperm motility appeared to decrease over time (p<0.001).

Conclusions

In a large sample of patients presenting to a single institution for fertility assessment, some aspects of semen quality declined across more than two decades. An understanding of the etiologies and driving forces of changing semen parameters over time is warranted.

INTRODUCTION

Infertility is a growing problem worldwide, affecting up to 12% of couples, and male factor specifically is present in up to 50% of cases [1]. The most basic test for the assessment of male factor infertility is a standard semen analysis, which reports numerous factors including semen volume, pH, concentration, total count, morphology, and motility. A confirmatory analysis must be repeated following an abstinence period of 2 to 7 days and at least 4 weeks apart [2]. Male infertility is therefore commonly attributed to abnormalities from these analyses and subsequently classified based on fertile reference ranges published by the World Health Organization (WHO) [3, 4, 5].

While some additional adjunctive tests do exist for the assessment of male fertility, the ongoing heavy reliance on semen parameters has prompted investigation as to how semen quality has changed over time. Speculation and controversy exist with respect to these temporal changes. Reports as early as the 1990s have detected a decrease in semen quality, while others during the same period have reported no significant decline [6, 7]. Data from more recent and larger study populations from Asia and Europe have similarly suggested worsening semen quality over time with respect to sperm concentration as well as other parameters, while other studies have demonstrated opposite conclusions, including one study suggesting an improvement in sperm count and concentration [8, 9, 10, 11, 12, 13]. The largest study assessing a more heterogenous population was a combined analysis of semen analyses from over 100,000 men from the United States and Spain, which showed a decline in semen quality. This study, however, only examined sperm count and concentration [14].

Given the ongoing controversy as to how semen quality is changing over time we elected to review data from our center. More specifically, our objective was to review temporal trends in multiple semen parameters in sub-fertile non-azoospermic men in a large tertiary fertility referral center to further understand semen quality changes over time.

MATERIALS AND METHODS

1. Study population

Data were retrospectively reviewed by the study team over a 20-year period from a prospectively collected electronic database between 2000 and 2019. Each consecutively retrieved semen analysis by the laboratory was completed in a written format and then transcribed to an electronic database stored on a secure workstation. All data were entered by a laboratory certified andrologist. Patients providing semen samples were referred from not only within New York City but also nationally and internationally, by male reproductive urologists, general urologists, reproductive endocrinologists, and/or primary care physicians for fertility evaluation. Samples with no sperm present (azoospermia) or known post-vasectomy samples were excluded from the cohort.

2. Ethical approval

Study approval was obtained by the Weill Cornell Medicine Institutional Review Board (approval number: 20-04021763). Informed consent was waived by the board owing to the restrospective design of the study.

3. Demographic data

Patient date of birth was recorded at the time of the patient visit, which was used to calculate age at the time of sample calculation. Time of collection as well as completeness of sample was recorded electronically. Method of collection was recorded, and abstinence periods were documented in days. Samples were grouped based on date of collection.

4. Sample collection and preparation

All patients were given instruction by their referring physician or male infertility specialist on specimen collection. Samples were to be obtained by masturbation into a sterile container without the use of spermotoxic lubricants. Samples were provided to the laboratory immediately after collection and were processed by a certified laboratory technician.

All single individual samples were manually prepared by a trained andrologist as per the 4th Edition of the WHO laboratory manual [5]. All andrologists were certified under the American Association of Bioanalysts as well as certified as a Clinical Laboratory Technologist per state requirements. Fluid pH was documented using standard pH paper from VWR Chemical Products (Radnor, PA, USA). Volume was assessed using a volumetric pipette and recorded in milliliters. Sperm characteristics were assessed following smear preparation on a glass slide utilizing microscopic examination at 400× magnification. Sperm counts were assessed and recorded in millions, and sperm concentration was reported as millions of sperm per milliliter. Motility was reported as a percentage of motile sperm visualized on the smear, and progressive motility was reported as grade 1 (sluggish with no progressive movement), grade 2 (slow, some forward progression), grade 3 (moving in straight line with moderate speed) or grade 4 (movement in straight line at high speed). Sperm morphology was recorded as a percentage of sperm classified as morphologically normal following assessment of adequate sperm heads, midpieces and tails. Strict morphology was not presented.

5. Statistical analysis

Statistical analyses were performed to examine temporal trends in all semen parameters both descriptively and quantitatively. Mean values and associated standard deviations were calculated for pH, progressive motility and motility as they appeared to have a normal distribution. Medians and interquartile ranges (IQRs) were reported for all other parameters as they did not appear to have normal distributions. Parameters with missing data were excluded and therefore complete-case analysis of each individual variable was performed. Daily data was aggregated at the month using the mean. Generalized least squares (GLS) was then used to estimate trends in average-monthly sperm quality measurements (volume, pH, concentration, total sperm count, motility, morphology, and progressive motility) from 2000 to 2019 adjusting for average-monthly age and abstinence days. Autocorrelation of residuals was accounted for. Non-linearities were modeled using restricted cubic splines with six knots. Wald tests of non-linear terms were performed to assess if the use of restricted cubic splines were appropriate. Contrasts of the estimated average sperm quality measurements based on GLS models were performed between the first month and last months of data collection. Sensitivity analysis were completed repeating analysis using only first time semen analysis data. Statistical significance was evaluated at the 0.05 alpha level and model estimates are presented with 95% confidence intervals. Analyses were performed using Stata v14 (Stata Corp, College Station, TX, USA) and R version 3.6.3.

RESULTS

A total of 10,865 semen analyses were reviewed by a single laboratory over the study period. After excluding those with a sperm concentration of zero, a total of 8,990 samples remained over a total of 20 years (Fig. 1). Data were missing for <0.01% of patients for volume, concentration, and total sperm count, motility, and morphology, but 7.0% (n=664) for pH.

Fig. 1 Consort diagram demonstrating included semen analyses.

1. Demographics

Demographic data for the entire cohort (n=8,990), including those with some missing semen parameter data, are summarized in Table 1. The median age for the complete cohort was 37.0 years (IQR, 33.0–43.0 y) and was similar amongst yearly groups between 36 and 39 years of age. Overall, median abstinence period was 3.0 days (IQR, 2.5–4.0 days) for the entire cohort, and was between 3 and 4 days for all time periods.

Table 1 Patient demographics and semen parameters based on yearly time periods between 2000 and 2019

2. Semen parameter trends

Semen parameters based on years are shown in Table 1. Given the non-linearity of the data, restricted cubic spline modeling was used to model the data graphically based on monthly averages (Fig. 2). A significant downward trend is demonstrated for volume, with the most pronounced changes between 2005 and 2010. Semen pH displayed a variable upward trend towards a more basic pH. Sperm morphology appeared to display a significant decline over time. For both sperm count and concentration, a peak is visualized between 2002 and 2006, otherwise the quality appears relatively unchanged over time. Both progressive motility and motility appeared relatively stable over the study period. Table 2 contrasts the average estimated outcomes at the beginning and end of the study period. Only semen volume, motility, and morphology were significantly different over time (p<0.001).

Fig. 2 Restricted cubic splines for all semen parameters between 2000 and 2019. (A) Semen volume (mL). (B) Semen pH. (C) Sperm morphology (%). (D) Sperm concentration (millions/mL). (E) Total sperm count (millions). (F) Sperm motility (%). (G) Sperm progressive motility (1–4).

Table 2 Difference in estimated average outcomes at end of data collection vs. beginning of data collection

DISCUSSION

We demonstrated that some semen parameters including volume, motility, and morphology, are worsening over time, while sperm count and concentration appear stable in a large heterogeneous population of subfertile non-azoospermic men. These findings support previous studies which concluded that semen quality may be declining or remains unchanged, but refutes those which suggest quality is improving [8, 9, 10, 11, 12, 13, 14].

Sperm count and concentration appear unchanged over the study period. However, there was a distinct peak in sperm concentration and total sperm count between the periods of 2002 and 2006. No laboratory specific factors or technical elements including sperm processing and assessment were identified that could account for our findings. However, an amendment to New York state law regarding improved insurance coverage for fertility treatment in women occurred in 2002, which may explain the influx of more fertile male partners seeking care while their partners were being evaluated [15].

As seen in previous studies, semen volume significantly decreased over time, while semen pH increased over time, but not this was not statistically significant [16]. Technical factors such as inadequate collection would not explain this change over such a significant time period. Abstinence period is known to influence semen volume, but this was adjusted for in our study [17]. A hypothesis to explain this observation may include increased rates of partial obstruction in some portion of the male reproductive tract from iatrogenic procedures, infections, or idiopathic sources [18]. Additional studies have also explored the impact of seminal vesicle characteristics and ejaculate volume. Reduced seminal vesicle size or abnormal seminal vesicle ejection fraction may provide an alternative explanation for our findings [19].

Sperm morphology significantly decreased over time, which is consistent with previously reported studies [20, 21]. During the study period there were two changes to the WHO criteria for sperm morphology. However, these changes were only modifications to the normal cut-off values and not changes to the actual criteria for which morphology assessment was completed [22]. Given the subjectivity in sperm morphology assessment, amendments to technical assessment over time could explain the discrepancies, but in our study this change was consistent despite having four different andrologists over the study period. However, it has been suggested that the introduction of new criteria over time may have caused a classification drift as technicians became more critical of their morphological assessments [21].

Sperm motility did decrease significantly over time. Previous studies have suggested that sperm motility is decreasing for men with normal sperm counts, but increasing for those with low counts [14]. Progressive motility in our study was generally reported as a range (i.e. 2 to 3) and therefore it was captured as an average of these values. Given the predominant reporting as a range as opposed to discrete values, this parameter should be interpreted with caution.

Various potential etiologies have been proposed to explain worsening semen quality over time, of which only age was included in our study. Aside from technical errors, increasing age has been shown to have an inverse association with total sperm motility, progressive motility, normal sperm morphology, and sperm concentration [23]. Other factors have been hypothesized, which were not included in our study, but may explain some of our associations. Obesity for example, is associated with reduced sperm morphology and sperm concentration [23]. Environmental factors such as exposure to pollutants including methyl mercury, pesticides, organic solvents, radiation, endocrine disrupting compounds, and even mobile phones have also been hypothesized to compromise male reproductive function [24, 25, 26]. Other factors include smoking and alcohol, and even cell phone usage [27].

The findings of our study do have important clinical implications. There is increasing evidence of an association between male infertility and systemic disease, including malignancies, autoimmune conditions, and high risk behavior [28, 29]. There is also evidence of worsened overall health with worse semen parameters [30]. If these trends are accurate, this overall suggests the need for an increased focus on a thorough history and physical evaluation in infertile men, and that if certain parameters are worsening perhaps this is reflective of additional previously undiagnosed medical disease, especially in young men who may otherwise not seek medical care.

There are several limitations to our study. Our retrospective study design limits obtaining any additional demographic or environmental data for analysis, introducing potential misclassification bias from erroneous data entry, and inability to obtain missing data. However, our study had a large sample size over a significant time period, so although missing data was prevalent for almost every parameter, the missing data was <1% for all but one parameter (pH), suggesting minimal impact from the missing data. Our study also includes all men obtaining semen samples for infertility workup, which may lend itself to selection bias, and did not discriminate referrals from primary care physicians, reproductive endocrinologists, general urologists, or reproductive urologists. An increased number of referrals from male reproductive urologists may potentially explain the decrease in semen parameters as they may represent more severe cases. Our study did not use the same andrologist over the entire study period but instead included four separate andrologists for various periods of time. While this suggests a potential role for observer bias, the timing of change in andrologists did not impact or correlate with any trends presented in our study. Finally, duplicate samples from the same patient were not removed from the study. However, sensitivity analysis (Supplement Table 1, Supplement Fig. 1) completed including only men with first time semen analysis provided similar results and also suggests additional statistically significant differences in total sperm count and pH.

CONCLUSIONS

In a large subset of non-azoospermic sub-fertile men presenting to a single institution for fertility assessment at a single high-volume laboratory, some aspects semen quality has significantly declined over the past twenty years but sperm concentrations appear unchanged. Further research exploring the etiologies and driving forces impacting altered semen quality over time are warranted.

Supplementary Materials

Supplementary materials can be found via https://doi.org/10.5534/wjmh.210201.

Supplement Table 1

Difference in estimated average outcomes at end of data collection vs. beginning of data collection for only first patient visit

Click here to view.(65K, pdf)

Supplement Fig. 1

Restricted cubic splines for all semen parameters between 2000 and 2019 for first visit only. (A) Semen volume (mL). (B) Semen pH. (C) Sperm morphology (%). (D) Sperm concentration (millions/mL). (E) Total sperm count (millions). (F) Sperm motility (%). (G) Sperm progressive motility (1–4).

Click here to view.(3M, pdf)

Notes

Conflict of Interest:The authors have nothing to disclose.

Funding:NP is supported by the Frederick J. and Theresa Dow Wallace Fund of the New York Community Trust.

Author Contribution:

Conceptualization: NP, SJK, MG.

Data curation: NP, OAHA, SJK, CAS.

Formal analysis: NP, NW.

Investigation: NP, SJK, MG.

Methodology: NP, MG.

Supervision: MG.

Visualization: NW.

Writing – original draft: NP, MG.

Writing – review & editing: All authors.

Data Sharing Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

1. Kumar N, Singh AK. Trends of male factor infertility, an important cause of infertility: a review of literature. J Hum Reprod Sci 2015;8:191–196.

2. Blickenstorfer K, Voelkle M, Xie M, Fröhlich A, Imthurn B, Leeners B. Are WHO recommendations to perform 2 consecutive semen analyses for reliable diagnosis of male infertility still valid? J Urol 2019;201:783–791.

3. Wang C, Swerdloff RS. Limitations of semen analysis as a test of male fertility and anticipated needs from newer tests. Fertil Steril 2014;102:1502–1507.

4. Cooper TG, Noonan E, von Eckardstein S, Auger J, Baker HW, Behre HM, et al. World Health Organization reference values for human semen characteristics. Hum Reprod Update 2010;16:231–245.

5. World Health Organization (WHO). WHO laboratory manual for the examination of human semen and sperm-cervical mucus interaction. 4th ed. Cambridge: Cambridge University Press; 1999.

6. Fisch H, Goluboff ET, Olson JH, Feldshuh J, Broder SJ, Barad DH. 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.

7. Auger J, Kunstmann JM, 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.

8. Li WN, Jia MM, Peng YQ, Ding R, Fan LQ, Liu G. Semen quality pattern and age threshold: a retrospective cross-sectional study of 71,623 infertile men in China, between 2011 and 2017. Reprod Biol Endocrinol 2019;17:107

9. Mishra P, Negi MPS, Srivastava M, Singh K, Rajender S. Decline in seminal quality in Indian men over the last 37 years. Reprod Biol Endocrinol 2018;16:103

10. Basnet P, Hansen SA, Olaussen IK, Hentemann MA, Acharya G. Changes in the semen quality among 5739 men seeking infertility treatment in Northern Norway over past 20 years (1993–2012). J Reprod Biotechnol Fertil 2016;5:1–7.

11. Sengupta P, Nwagha U, Dutta S, Krajewska-Kulak E, Izuka E. Evidence for decreasing sperm count in African population from 1965 to 2015. Afr Health Sci 2017;17:418–427.

12. Levine H, Jørgensen N, Martino-Andrade A, Mendiola J, Weksler-Derri D, Mindlis I, et al. Temporal trends in sperm count: a systematic review and meta-regression analysis. Hum Reprod Update 2017;23:646–659.

13. Jørgensen N, Joensen UN, Jensen TK, Jensen MB, Almstrup K, Olesen IA, et al. Human semen quality in the new millennium: a prospective cross-sectional population-based study of 4867 men. BMJ Open 2012;2:e000990

14. Tiegs AW, Landis J, Garrido N, Scott RT Jr, Hotaling JM. Total motile sperm count trend over time: evaluation of semen analyses from 119,972 men from subfertile couples. Urology 2019;132:109–116.

15. National Conference of State Legislatures. State laws related to insurance coverage for infertility treatment [Internet]. Washington, D.C.: National Conference of State Legislatures; c2019 [cited 2020 Jun 14]. Available from: https://www.ncsl.org/research/health/insurance-coverage-for-infertility- laws.aspx.

16. Carlsen E, Giwercman A, Keiding N, Skakkebaek NE. Evidence for decreasing quality of semen during past 50 years. BMJ 1992;305:609–613.

17. Ayad BM, Horst GV, Plessis SSD. Revisiting the relationship between the ejaculatory abstinence period and semen characteristics. Int J Fertil Steril 2018;11:238–246.

18. Baker K, Sabanegh E Jr. Obstructive azoospermia: reconstructive techniques and results. Clinics (Sao Paulo) 2013;68 Suppl 1:61–73.

19. Lotti F, Corona G, Colpi GM, Filimberti E, Innocenti SD, Mancini M, et al. Seminal vesicles ultrasound features in a cohort of infertility patients. Hum Reprod 2012;27:974–982.

20. Rolland M, Le Moal J, Wagner V, Royère D, De Mouzon J. Decline in semen concentration and morphology in a sample of 26,609 men close to general population between 1989 and 2005 in France. Hum Reprod 2013;28:462–470.

21. Morbeck DE, Leonard PH, Weaver AL, Shimek KM, Bouwsma EV, Coddington CC. Sperm morphology: classification drift over time and clinical implications. Fertil Steril 2011;96:1350–1354.

22. Menkveld R. Clinical significance of the low normal sperm morphology value as proposed in the fifth edition of the WHO laboratory manual for the examination and processing of human semen. Asian J Androl 2010;12:47–58.

23. Tsao CW, Liu CY, Chou YC, Cha TL, Chen SC, Hsu CY. Exploration of the association between obesity and semen quality in a 7630 male population. PLoS One 2015;10:e0119458

24. Foster WG, Neal MS, Han MS, Dominguez MM. Environmental contaminants and human infertility: hypothesis or cause for concern? J Toxicol Environ Health B Crit Rev 2008;11:162–176.

25. Mínguez-Alarcón L, Hauser R, Gaskins AJ. Effects of bisphenol A on male and couple reproductive health: a review. Fertil Steril 2016;106:864–870.

26. Jurewicz J, Hanke W, Radwan M, Bonde JP. Environmental factors and semen quality. Int J Occup Med Environ Health 2009;22:305–329.

27. Virtanen HE, Jørgensen N, Toppari J. Semen quality in the 21st century. Nat Rev Urol 2017;14:120–130.

28. Eisenberg ML, Li S, Cullen MR, Baker LC. Increased risk of incident chronic medical conditions in infertile men: analysis of United States claims data. Fertil Steril 2016;105:629–636.

29. Cedars MI, Taymans SE, DePaolo LV, Warner L, Moss SB, Eisenberg ML. The sixth vital sign: what reproduction tells us about overall health. Proceedings from a NICHD/CDC workshop. Hum Reprod Open 2017;2017:hox008

30. Eisenberg ML, Li S, Behr B, Cullen MR, Galusha D, Lamb DJ, et al. Semen quality, infertility and mortality in the USA. Hum Reprod 2014;29:1567–1574.