Theses Doctoral

A life course approach to potentially modifiable risk factors for poor semen quality

Kahn, Linda Gross

Poor semen quality is an indicator of male infertility and is also associated with a variety of adverse health outcomes in men. It is therefore important from both clinical and public health perspectives to discover predictors of poor semen quality, especially those that are potentially modifiable. My dissertation research focuses on two of these potential risk factors: adiposity and stress. Unlike most studies to date, which have only considered the relationship between these exposures and semen quality cross-sectionally, my research takes a life course approach. I explore associations between birth weight, adiposity in both childhood and adulthood, and allostatic load—a construct representing the effect of cumulative stress on the body’s regulatory systems—and three commonly-used semen outcome parameters: sperm concentration, percent progressive motility, and percent normal morphology. The logic that underlies this approach is that while sperm are constantly being produced from sperm stem cells in the testes, which would argue in favor of cross-sectional studies, the sperm stem cells themselves and the Sertoli and Leydig cells that stimulate and nurture that metamorphosis are laid down in the fetal period and undergo important developmental and proliferative phases during early childhood and puberty that affect their number and functional health in adulthood.
Using data from 193 participants in the Study of the Environment and Reproductive health follow-up to the Child Health and Development Studies birth cohort, I was able to calculate birth weight for gestational age percentile (bw/ga) and six age-appropriate adiposity measures (at 4 months, 12 months, and 4 years, and in participants’ 20s, 30s, and at the time of semen collection), then test for their independent, critical period, and cumulative associations with the three semen outcomes as well as a combined outcome measure of subfertility based on World Health Organization reference levels. While bw/ga was positively associated with sperm concentration, subsequent childhood adiposity measures showed increasingly negative associations, and none of the adult measures were significantly associated with concentration. By contrast, only adult measures were associated with motility and morphology. This suggests that there may be critical periods in childhood when adiposity negatively affects sperm concentration by interfering with the development and proliferation of Sertoli and Leydig cells. Accumulation of oxidative stress in the testes due to overweight/obesity may explain the negative relationships between adult adiposity and sperm motility and morphology.
To investigate allostatic load’s relationship to semen quality, I conducted a pilot study at Columbia University’s Center for Women’s Reproductive Care that enrolled 61 men who were having their initial diagnostic semen analysis and blood draw on the same day. Blood samples were analyzed for 7 biomarkers associated with homeostatic regulation across several physiologic domains. I then created an allostatic load scale in which participants were assigned 1 point for being in the high-risk quartile for systolic blood pressure, diastolic blood pressure, body mass index, or any of the biomarkers. In regression analyses, allostatic load was not associated with either sperm concentration or morphology, but showed an unexpected positive association with motility. This association was entirely driven by the six participants who scored 0 on the allostatic load scale and who did not differ from the rest of the sample in any way that could plausibly be linked to reduced motility. I therefore concluded that this was a spurious finding. In further analysis of the allostatic load variable itself, I found that few of its individual components were correlated with the semen outcomes. This contrasts with other studies of allostatic load and adverse health outcomes, but these have generally been conducted in either elderly or stressed populations, neither of which described my cohort. Allostatic load may not be a reliable measure of stress in reproductive age populations and may not capture regulatory systems appropriate to reproductive health outcomes.
My dissertation highlights the value and challenges of conducting semen quality research from a life course perspective. Future studies should consider collecting longitudinal data on adiposity and stress, as well as repeated semen samples beginning in adolescence in order to further our understanding of the natural progression of semen quality across the reproductive life span and provide the opportunity to explore whether modifying these risk factors affects semen quality.


  • thumnail for Kahn_columbia_0054D_13701.pdf Kahn_columbia_0054D_13701.pdf application/pdf 7.13 MB Download File

More About This Work

Academic Units
Thesis Advisors
Factor-Litvak, Pam
Ph.D., Columbia University
Published Here
January 4, 2017