Around the world in pursuit of (vanishing) sperm
Author: Jonathan Galka
I’m departing Holešovice Station in Prague and heading northwest on the Arriva bus #580916. The skyline fades quickly from concrete residential blocks into mixed deciduous forest and farmland. Past the airport, the city of Kladno comes into view. Visible now are the buildings which once comprised Poldi Steel Factory, Central Bohemia’s largest employer until the company was privatized and divided. Kladno is described as the birthplace of heavy industry in Central Bohemia. Its mines were built in 1842 and worked by Jewish prisoners interned in Theresienstadt (Terezin) Ghetto during WW2. The mines also allowed for the survival of the Central and Northern Bohemian industrial economy after the larger collapse of industry following the dissolution of the Eastern Bloc. Today, a LEGO factory rises from the landscape, built in 1999 as LEGO began moving production out of Denmark, first to Hungary and the Czech Republic and then to Mexico in 2009 and Zhejiang Province, China in 2015.
Continuing on my journey, some time after Kladno has faded again into mixed pastoral land and passing the towns of Žatec and Most, I disembark at my destination of Teplice. Teplice as a city has embodied the extremes of 20th-century Europe. Beginning as an Austro-Hungarian health destination (Teplice in Old Czech means hot springs), it witnessed the demographic and geopolitical consequences of interwar Sudeten German nationalism followed by German annexation and wartime Holocaust, culminating in the post-war banishment of its German speaking majority under the Beneš Decrees.
Teplice has been many places, but on an overcast winter day like today, back in the mid-1990s, Teplice was coping with being in the heart of “one of the most polluted places on Earth.” Winter temperature inversions caused industrial and domestic fumes to merge and blanket the valley. The population breathed air containing six to nine times the WHO-recommended quantity of sulfur dioxide, while a milieu of heavy metals, organics, and polycyclic aromatic hydrocarbons circulated in volumes exceeding their maximum recommended levels by up to a factor of 100. In 1995, endocrinologist Howard Kulin and andrologist Niels Skakkebaek concluded that the region “is a hostile place, indeed, where there have been many decades of abuse.”
Environmental toxicologists, along with a range of health scientists, saw Central and Northern Bohemia at this time as something of a laboratory; some researchers compared what could be learned from the Eastern European polluted regions to what was learned about the bodily impact of ionizing radiation on survivors in Hiroshima and Nagasaki. Capitalizing on the opportunity, Kulin, based in America, and Skakkebaek, in Denmark, put forward a series of bold new research programs centered around understanding the consequences of environmental deterioration for human health. They would focus especially on reproduction, with broad geographical purview, emphasizing “the need to enlarge world wide data collecting.”
One project, the Teplice Program, would leverage new U.S.-Czech collaborations to compare reproductive, neurobehavioral, and respiratory conditions and outcomes between Teplice and Prachatice, a relatively less polluted region in South Bohemia. Another study in Aarhus, Denmark, would be similar in scope and method, while focusing predominantly on Western Europe to illuminate the effects of occupational exposure to lead, certain pesticides, and styrenes on male reproductive health.
Kulin and Skakkebaek were anxious to empirically track what they perceived as rapid changes in men’s reproductive health and health outcomes across Europe. As Eastern Europe emerged from decades of harsh conditions, a shocking diminishment in health outcomes became apparent. As recently as the mid-1960s, life expectancy at birth in Czechoslovakia had come to nearly equal Western Europe, but by the 1980s this trend was in reverse. And this was a crisis with a distinctly male bent. Standardized for age, male mortality rates in the former Eastern Bloc countries were rising. Mortality rates for middle-aged men with heart disease in Czechoslovakia were double compared with Sweden or West Germany, and if you were a working-aged man, you had twice the chance of dying in the East as in the West. What’s more, some Western countries, namely Denmark, were showing a leveling-off of life expectancy as in the East, without the added twenty years of environmental stress of the East.
Kulin and Skakkebaek were careful not to overstate the evidence for environmental effects on male reproduction in Eastern Europe. Smoking, psychosocial stress, diet, alcohol use, and exercise were also likely very important, and while environmental pollution was also on the list, it was “likely not to the degree that may have been initially thought.” All the same, they suggested in the same 1995 paper that investigations into Eastern Europe would be an inroad into “the promotion of world wide programs which closely examine the relationship of the environment to potentially adverse reproduction outcomes.” Such broad understanding was imperative, given that “adverse environmental effects on reproduction may be a catastrophe about to happen.” In my view, a fair enough aspiration.
The remainder of this post will dive deeply into what can be gleaned from these early studies comparing sperm quality across geographical regions within Europe for contextualizing current high-profile claims about the decline of sperm around the world. Such claims are reflected in the 2017 study by Levine et al., the largest meta-analysis of sperm count papers to date (globally from 1973-2011), and a project that is very much the fruit of Kulin and Skakkebaek’s ambitions for research on the global distribution of sperm count. To ground this exploration, the Gender Science Lab worked with the Harvard Center for Geographic Analysis to map the more than 43,000 individual samples included in the 2017 study by Levine et al. (Map 1).
Map 1. Sample size per country in Levine et al. meta-analysis. Countries whose total sample is denoted by dark blue are categorized by the authors as “West” and those in green as “Other.”
Levine et al.’s operative claim is that sperm count (and so by proxy, male reproductive health) is declining at an unchecked rate among Western men, but such a trend cannot be found among men in the rest of the world, grouped and analyzed as the category “Other.” The goals of visualizing Levine et al.’s data were to understand the geographical and temporal distributions of sperm quality measurement in Levine et al.’s study period. How does the sampling of sperm vary among Levine et al.’s “West” and “Other” categories, and how does that variation emerge and change in time? From these exercises in mapping, we learned that the global composition of studies examining sperm count as represented by the 2017 meta-analysis is fundamentally geographically asymmetrical, an observation that gains importance in light of Levine et al.’s claims. As Map 1 shows, the metaanalysis reveals the huge preponderance of studies on male reproductive health from the West, especially Northwestern Europe, and the paucity of samples included from much of the rest of the world, Eastern Europe included. We also learned that this asymmetry between “Western” and “Other” studies is produced temporally. Studies conducted earlier in the period 1973-2011 were predominantly Western European, allowing for claims to be made about Western male decline against a background of mostly speculative “Other” males, on whom studies were undertaken principally later in Levine et al.'s period of analysis (time series map, sperm count by country).
On “West” and “Other”: Japan and India and Denmark
Grouping places together which have less in common over time than we might think has consequences. Levine et al.’s framing disappears distinctions within the “West,” and of course within “Other,” groups of places that have transformed radically during the study period and with appreciable consequences for reproductive health. A particular kind of ostensible male is constructed, and only this male can be called imperiled; only this male might benefit from desired research on (sub)fertility and its causes and implications.
Japan
This disappearing act works deleteriously within the category of “West” as it does across the “West”-“Other” categorical divide. Japan is a country that, demographically, might well have been included in Levine et al.’s “West” category, but was not. Niels Skakkebaek has begun to realize his ambitions of collecting global comparative data as he, with his colleague Niels Jørgensen in the Department of Growth and Reproduction at the Rigshospitalet in Copenhagen, developed research collaborations in Japan. Niels Skakkebaek and Niels Jørgensen are close collaborators, listed as co-investigators on all of each other's research projects since 2000, and they are among the most proliferative researchers working on the relationship between the environment, endocrine disruption, and reproductive (especially male) function. They both hold an MD from University of Copenhagen, while Jørgensen holds an additional medical PhD and an andrological certification from the European Academy of Andrology. Also, both Jørgensen and Skakkebaek are publically outspoken about the imperilment of men exposed to the barrage of chemical pollutants present in the modern world. According to his faculty profile, since 1996 Jørgensen has been involved in international projects investigating male reproductive health, and we have already seen that by 1995 Skakkebaek sought to develop international programs to the same end. However, aspirations of developing global comparative research seem unfulfilled when comparing the kinds of papers included by Levine et al. with which the Danish endocrinologists were involved. In the Levine et al. meta-analysis, papers from four “Western” states list Skakkebaek as a co-author, and the same is true of Jørgensen for papers about more than ten “Western” states and territories. However, Japan is the only country categorized by Levine et al. as “Other” whose papers list either Skakkebaek or Jørgensen as a co-author.
Two of the Japanese studies included in Levine et al.’s meta-analysis list the Danish researchers as co-authors. Akutsu et al (2008) looked at the effects of Polybrominated Diphenyl Ethers (PBDEs), a class of environmental pollutants and purported endocrine disruptors, on sperm quality. They found firstly that sperm serum concentrations of PBDEs were similar to those found in Western Europe, and less than those found in the U.S.A., and that concentration of only one out of 13 measured PBDE congeners (out of use in Japan since the 1990s) was associated with lower sperm concentration. Iwamoto et al (2013), meanwhile, sought to establish reference values for temporal analysis of sperm quality parameters across four Japanese cities. They found that among cities, while sperm concentration did not change, sperm morphology and motility did, and also that smoking (despite relative prevalence) had no effect on sperm quality. Among countries whose demographic and environmental characteristics align closely, the relationship triangulated between environmental pollution, lifestyle behaviors, and sperm count is far from clear.
India
To demonstrate that the category of “Other” does perhaps the most work at vanishing the contours of both male reproductive health and the global distribution of environmental toxicants, consider a comparison between the case of Japan and the case of India. India, a subcontinent of 1.35 billion inhabitants, spread across twenty-nine states and nine territories, is represented in Levine et al. (2017) by 1,500 individuals. Among these studies, Marimuthu et al. (2003) found no change in semen quality for 1,176 men in Delhi over eleven years, but Gopalkrishnan (2000) did find a significant decline in semen quality in Mumbai, and Mukhopadhyay et al. (2009) identified declining motility and volume among men in Kolkata, as Adiga et al. (2008) did in South India.
Levine et al. included within their analyses only unexposed controls from studies analyzing the effects of environmental exposures. However, the distinction between the exposed and unexposed condition blurs when one considers the pervasiveness of ambient and background environmental contamination present at high levels not only in factories and fields, but also in the air, water, and food consumed by much of the world’s population. This is exemplified by India, where researchers continue to assert that high levels of pollution and the ubiquity of endocrine disrupting chemical contamination contribute to general declines in fertility parameters across the country. And within India, the relationship between local environmental pollution and male reproductive health is complicated; Delhi performs worse than both Mumbai and Kolkata in relative air and drinking water quality by WHO standards, mismatching with sperm decline data.
“China and India, to say nothing of Central and South America, are enormous and diverse in terms of geography, demography, lifestyle and behavior, reproductive health outcomes; in short, every factor relevant to the construction and justification of Levine et al.’s analysis.”
What’s more, a systematic review by Mishra et al. (2018) covering all Indian regions found a significant decline in semen quality parameters, particularly sperm concentration and normal morphology, for a total of 13,486 Indian men between 1979-2016. A more robust and textured analysis of the subcontinent found precisely the outcome which Levine et al. apparently lacked the statistical power to find. Differences in environmental quality and demographics between much of urban Japan and India could not be more stark, and yet both states are collapsed into the category of “Other.” In effect, this collapsing disappears any texture to be found in the sperm count decline hypothesis.
Denmark
One more aspect of Levine et al.’s geospatial categorizing requires unpacking, and to do so we will depart India for Denmark.. Levine et al. differentiate between two types of samples in the studies they include in their meta-analysis: fertile men and “unselected” men (those whose fertility status is unknown). I will focus on studies of men unselected by fertility, as those men are assumed to be representative of the general population in a given geographic area. Nevertheless, as we will see, there is enormous variation in the number and type of studies of unselected men Levine et al. use to make claims about sperm decline in various parts of the world, which calls the validity of such claims into question.
Eleven studies of unselected men were included in Levine et al.’s meta-analysis from Denmark, including more than 10,000 men. This makes Denmark by far the most sampled country included in the 2017 meta-analysis (time series map, sperm count by city). Three studies sought to characterize sperm quality within the general population of Danish or Faroese men. Jørgensen et al. (2012) conducted a prospective cross-sectional study of the general Danish male population unselected by fertility, including nearly 5,000 men, and found that one in four men met the threshold for optimal semen quality. This is substantially lower when compared with historical data and with data from recent fathers. Gyllenborg et al. (2001) found that among unselected semen donors, sperm counts increased between 1977 and 1995, and that sperm count and concentration varied significantly by season. Meanwhile Halling et al. (2013) found in a cross-sectional study that Faroese men living in the Faroe Islands, a Danish territory in the North Atlantic, had similar sperm concentrations to Danish men in Denmark. A further two studies carried out by Jensen et al. (1997) and Bang et al. (2005) aimed to understand trends within specific subsets of the Danish male population, and two more by Kolstad et al. (1999) and Larsen et al. (1999) focused on the effects of occupational chemical exposure on sperm quality. Danish sperm quality research includes not only prospective studies of the Danish population, but also studies that examine general temporal and geographic variation in sperm parameters, as well as those that examine the effects of transient exposures on sperm within the specific Danish context. With the largest number of samples relative to population size in the 2017 meta-analysis, the presence of such data with relatively substantial ecological granularity suggests that trends might well be inferred for the state of sperm quality in Denmark.
Now, looking at the LEGO factory in Billund, I notice how similar it looks to the factory back in Kladno; this is no coincidence. LEGO designed modular building units to be re-assembled in pre-configured designs the world over. The factories in Monterrey, Mexico, and in Zhejiang Province will trigger the same sensation of deja vu. The sameness has been engineered, distributed, and increasingly exported beyond Western Europe, to Eastern Europe, to Central and South America, and to East Asia.
Unfortunately, the granularity found in Danish studies of sperm quality is very far from replicated elsewhere in the world. In Mexico, contrastingly, no studies on unselected men were included in the 2017 meta-analysis that sought to characterize the sperm quality of the general population, or of a subset of the population. Rather, a total of 58 men from two studies were either factory or agricultural workers. In one study, unexposed men were office workers in a rubber factory apparently unexposed to the volatile hydrocarbons wafting about the production floor, while in the other, unexposed agricultural workers only helped prepare the organophosphorus pesticide solutions sprayed by their exposed counterparts. Three out of five studies included from the entirety of Central and South America were studies on occupational exposure. From Zhejiang Province, meanwhile, 37 men were included from a study by Xiao et al. (2001) that examined the effects of occupational exposure to benzene, toluene, and xylene on sperm quality; the study identified significant effects on several sperm quality parameters. The population of Zhejiang Province alone is north of 57 million, nearly ten times that of the state of Denmark, and that of China more than 220 times that of Denmark. The starkness of these population differences relative to the inclusion of samples in the 2017 meta-analysis is thrown into sharper relief when Map 1 is further disambiguated by city and region (Map 2).
Map 2. Sample size per city (or region, where city was not available) in Levine et al. meta-analysis.
Owing to a combination of low statistical power and an absence of includable studies, Levine et al. found no significant declines in sperm count across Asia writ large. Contrastingly, Huang et al. (2017) found among 30,636 sperm bank donors in Hunan Province a significant decline in sperm count and concentration over fifteen years. China contains 56 recognized ethnic groups in 34 geographic provinces. Li et al. (2009) emphasized that across China, large proportions of men have abnormal semen parameters by WHO reference criteria, but that these proportions vary significantly across geography after adjusting for confounders, which points to wide variability across China in regional exposure to pollution, lifestyle changes, and ethnic admixture.
These are only a few examples I’ve pursued on my bus ride to Teplice. China and India, to say nothing of Central and South America, are enormous and diverse in terms of geography, demography, lifestyle and behavior, reproductive health outcomes; in short, every factor relevant to the construction and justification of Levine et al.’s analysis. Consequently, the discrepancy between the under-representation of some men in the meta-analysis, relative to the global population and its diversity and over-representation of others (see Maps 1 & 2), raises several important questions.
“West” and “Other” are not helpful categories in parsing the relationship between environment, lifestyle, and reproductive health outcomes like sperm quality parameters. As we have seen, environmental pollution and lifestyle behaviors do not map well onto this geographical construction, nor are their relationships to sperm quality parameters like sperm count well understood. These complexities, compounded over time, blur the environmental and demographic boundaries of so-called Western and non-Western countries, and the divide between them, to the point of incomprehension.
Disappearing Geographies and Future Directions
In retrospect, what transpired from the male reproductive studies undertaken by the Teplice Program? Papers published in 2000 and 2005 built upon the preliminary data available to Kulin and Skakkebaek in 1995 to suggest that periods of elevated air pollution in the Czech Republic likely altered the chromatin structure of sperm DNA, while leaving sperm count and volume unaffected. And what of the Aarhus program, which centered on Denmark (by far the most thoroughly-studied country with regard to sperm quality) to explore the effects of occupational exposure to heavy metals and plastics on male reproductive health? Only two out of eleven studies on unselected men conducted in Denmark that are included in the Levine et al. meta-analysis assessed occupational exposure.
Twenty-five years later, Kulin and Skakkebaek’s goals seem less than fully realized. Their evidence for differences in health outcomes between Eastern and Western Europe (differences that persist today), and perhaps genuinely concerning evidence around DNA damage related to cyclical air pollution exposure in Eastern Europe resulting from the Teplice Program (as just one example) have been collapsed into Levine et al.’s premises that geopolitical categories can be treated as homogeneous over the study period (1973-2017), and that the “canary in the coal mine for male reproductive health across the lifespan” can be understood as sperm count.
Levine et al. (2017) presented the categories of “West” and “Other” as coherent groupings of places with shared demographic and environmental characteristics between 1973 and 2011. The authors conclude by gesturing to possible explanations for their observed Western decline, mentioning that sperm count has been “plausibly associated with multiple environmental and lifestyle influences,” including prenatal maternal smoking and endocrine disrupting chemical exposure, and lifestyle alterations and pesticide exposure as an adult.
Given the finding that “Western” sperm count is in decline whereas “Other” sperm is not, the implication is that Western men are facing these influences more so than non-Western men. This not only takes for granted relationships among environment, lifestyle, and reproductive health that have texture well beyond “West” and “Other,” but it also disappears what we do know about the shifting global distribution of pollutants, toxicants, and lifestyle behaviors over time. And such a disappearance is nontrivial. It matters which males in which places are positioned as imperiled, and which are not.
What, then, do the categories “West” and “Other” achieve for Levine et al., and for the sperm decline hypothesis? They achieve the artificial construction of two kinds of men. One is decidedly imperiled, and his declining sperm count is the canary-song of a catastrophic near future. The other may or may not be experiencing a decline; a lack of statistical power and relevant studies for inclusion disables a stronger position on the subject. Non-Western decline notwithstanding, this man serves as a foil to the reproductive future of the Western man, no matter how strongly the evidence suggests that the very same environmental stressors triggering the reproductive anxieties of the Western man actually affect the “Other” man just the same, if not much more.
If “the promotion of worldwide programs which closely examine the relationship between the environment and potentially adverse health outcomes” was a “logical future next step” to Kulin and Skakkebaek in 1995, then it seems such a goal has been left unrealized, an absence leaving potential consequences for men the world over.
“Until such an approach propounded by the SCB can be enacted, relationships among geographical, toxicological, developmental, and behavioral variables and sperm quality, as well as their consequences for the imperilment of certain male bodies, remain questions to be answered.”
A sperm count biovariability hypothesis (SCB), as we propose in “The Future of Sperm: A Biovariability Framework for Understanding Global Sperm Count Trends,” published this week in the journal Human Fertility, is conscious of the complexity of these variables in their interactions and their potential consequences. SCB is committed to an understanding of sperm quality parameters as variable within a wide and non-pathological range in response to both endogenous and exogenous stimuli on both transient and developmental scales. This species-specific variability inbuilt within sperm quality parameters is present at the scale of the individual as well as the population. SCB is also committed to research on the interactions between geography, environmental pollutants and toxicants, lifestyle behaviors, and sperm quality parameters that approaches such interactions through confounder-controlled prospective studies engaging repeat individual measures. SCB recognizes at once that sperm count, and other sperm quality features, varies across bodies, ecologies, and time periods, and also that the reproductive health of some men in some places might well be shifting in detectable patterns, perhaps for the worse. SCB endeavors to provide a framework for the study of sperm quality without oversimplifying geographical categories or overdetermining the relationship between place and body.
As I said earlier, Teplice in the Czech Republic has been many places. Once it was the heart of the Central Bohemian Coal Mining District. Standing in the town square today, looking out towards the cathedral of St. John the Baptist, it’s hard to imagine that 25 years ago, this place ranked among the most polluted on Earth. I wonder where in the world shared this toxicological misfortune then, and where in the world people are subjected to similar circumstances now? As the production work of manufacturing and agriculture is increasingly exported beyond the boundaries of Levine et al.’s Western grouping, so too are many forms of pollution; following the trail of LEGO tells one cursory tale.
What this post has sought to suggest is that in geographical categories, especially in those capacious enough to be called “West” and “Other,” there is a plurality of places, experienced in turn by different people at different times. “West” and “Other” are not tenable categories, and an analysis of their boundaries from multiple angles sees their geographic and demographic architectures crumble. Europe in 1995 versus 2017, Zhejiang versus Hunan, or manufacturing work in Denmark versus Mexico, are all experienced and embodied differently. Rigorously attending to these differences is the only way to approach questions around the relationship between geography and male reproductive health. Until such an approach propounded by the SCB can be enacted, relationships among geographical, toxicological, developmental, and behavioral variables and sperm quality, as well as their consequences for the imperilment of certain male bodies, remain questions to be answered.
Recommended Citation
Galka, J. “Around the world in pursuit of (vanishing) sperm.” GenderSci Blog. 2021 May 4, genderscilab.org/blog/vanishing-sperm-geographies
Statement of intellectual labor:
Jonathan Galka authored this piece and all views expressed are his. Sarah Richardson, Marion Boulicault, Annika Gompers, and Kelsey Ichikawa offered helpful guiding suggestions and contributed substantial edits to earlier drafts.
Blog post thumbnail image is credited to Jan Pedersen.