2D: Insights from Past Research


Agent Orange herbicide was heavily used by the US military in Vietnam, resulting in large-scale exposure for local communities and veterans alike.  The resulting health problems, including birth defects, led to an increase in research on the reproductive impacts of herbicides.

A number of studies have focused on populations exposed to herbicides, in an effort to detect linkages between herbicide use and exposure and adverse birth outcomes.

The Agent Orange herbicide sprayed in Vietnam during the war is one, widely studied example of such linkages. Agent Orange was a 50%-50% mixture of two phenoxy herbicides – 2,4,5-T and 2,4-D. Its heavy use to defoliate the jungle in Vietnam led to exposures among U.S. military personnel, as well as Vietnamese living in rural areas.

Concern over lingering health effects in people exposed to Agent Orange led the EPA to fund a series of studies on the potential reproductive impacts of 2,4-D and other phenoxy herbicides in the U.S.

A series of hospital and birth-center based research projects were carried out from the early 1990s through 2005 that focused on the impacts of widely used herbicides, and especially 2,4-D and other phenoxy herbicides, on reproductive outcomes and birth defects in Wisconsin, Minnesota, and the Dakotas. Comparable studies were carried out in Canada.

The resulting body of work reported consistent associations between springtime herbicide applications and failure to conceive, spontaneous abortion, and birth defects.

Glyphosate, 2,4-D, and dicamba were among the herbicides identified in this body of research as increasing the risk of adverse reproductive outcomes. As evident in the data presented in the Herbicide Use section, the use of these three herbicides has increased as a result of heavy reliance on GE-HR crops, coupled with the spread of glyphosate-resistant weeds. Additional and significant increases — especially in 2,4-D and dicamba use — are bound to occur in the next three to five years.

Health Risks

Reproductive impacts such as failure to conceive and spontaneous abortion have been linked to herbicide exposure.

A large body of evidence links herbicide exposures to modest to moderately heightened risk of adverse reproductive outcomes, including failure to conceive and spontaneous abortions, birth defects, and developmental abnormalities (for a review, see Roberts and Karr, 2012).

In many studies, adverse reproductive outcomes and birth defects spike in heavily sprayed rural areas among women whose first trimester of pregnancy overlaps with the spring, herbicide spray season (Arbuckle et al., 2001; Chevrier et al., 2011; Garry et al., 1996, 2002a and 2002b; Rappazzo et al., 2016; Schreinemachers, 2003 and 2010; Winchester et al., 2016).

A study carried out in Indiana reported a significant association between the frequency of birth defects in 11 of 22 categories studied and the time period when pesticide levels (mostly herbicides) were highest in surface water (Winchester et al., 2009).

A birth center study was carried out in four wheat-producing states where 2,4-D and another phenoxy herbicide (MCPA) were heavily relied upon by farmers (29% of national wheat acres treated with 2,4-D in 2012, 34% with MCPA (USDA, Quick Stats)). Schreinemachers et al. (2003) reported higher rates of circulatory/respiratory malformations (OR=1.65) (all Odds Ratios reported are statistically significant, p=0.05). Infants conceived during April-June also had more circulatory/respiratory malformations than infants conceived at other times (OR=1.75). Male babies were at heightened risk of death from congenital anomalies (OR=2.66).

A similar study in Canada observed a moderate increase in risk of early abortions following exposures to phenoxy herbicides (e.g., 2,4-D and dicamba; OR=1.5) (Arbuckle et al., 2001). Preconception glyphosate exposure was linked to late abortions (weeks 12-19; OR=1.7), as were post-conception exposures (OR=1.4).

Women who conceived at age 35 or older faced 2-X to 4-X higher risk of a number of adverse birth outcomes across all herbicides, the phenoxy acid class, and glyphosate (Arbuckle et al., 2001). Based on these results, Arbuckle et al. concluded that the window of heightened risk for early spontaneous abortions (<20 weeks) is 3 months prior to conception through the first month of pregnancy.

The impact of spring herbicide and other pesticide applications on birth defects were studied in cohorts of pesticide applicators and farm families in Minnesota’s Red River Valley. The team concluded that spring herbicide applications in intensively farmed regions might be a risk factor for a number of birth defects, as well as springtime miscarriages (Garry et al., 1996 and 2002b).

In addition, this study supports the need for follow up of live births in order to fully capture the incidence of birth defects. Only 63% of the total number of birth defects was diagnosed in the first year of an infant’s life, and 26% were diagnosed at age 3 or later (Garry et al., 2002b).

Alarmingly, many birth defects are not identified until children are 3 years old or older. Source: Garry et al., 2002b

Significantly, the Garry et al. team reported the highest odds ratio across all associations for glyphosate and neurobehavioral disorders (specifically, Attention Deficit Disorder (ADD/ADHD); OR=3.6).

Glyphosate was applied by one or both of the parents of 6 children with ADD/ADHD, out of a total 14. It is worth noting that this association occurred during a period of time before the widespread planting of GE-HR crops.

In 2016, a team of French scientists published a paper on another key herbicide associated with GE-HR crops – glufosinate ammonium. This broad-spectrum herbicide is better known as Liberty, and is the herbicide associated with Liberty Link crops. The team delivered glufosinate intranasally to pregnant mice, at a dose ~5 to 25 times lower than the EPA-set, “acceptable” doses.

Large differences were recorded in brain and neurological system development in the offspring from the glufosinate-treated female mice. There was nearly a 50% increase in the thickness of the sub-ventricular zone (SVC) in the brain, a change that has been linked to a range of neurodevelopmental disorders, including Autism Spectrum Disorder (Herzine et al., 2016).

This complex study reported both changes in the production of neurons within the brain and changes in the migration of newly produced neurons within the brain. The team also reports that glufosinate is clearly among those environmental toxicants that can cause more serious developmental anomalies at low doses, compared to higher ones.

A study of 304,906 live births in North Carolina compared the incidence of birth defects as a function of proximity to fields treated with pesticides, of which herbicides were the most common pesticide applied. Investigators reported elevated ORs (1.98) for congenital heart defects and certain GI, genitourinary, and musculosketal abnormalities (Rappazzo et al., 2016).

Rocheleau et al. (2006) conducted a meta-analysis across nine qualifying studies and showed a link between maternal pesticide exposure and hypospadias (Pooled Risk Ratio=1.36), despite potential exposure.

Herbicides and Epigenetic Risks

This seminal article in 2005 by Anway et al. published in Science was the first to explore the impact of epigenetic changes.

Over the last decade-plus, scientists have discovered a cluster of new genetic mechanisms that can lead to developmental anomalies and heighten the risk of adult-onset chronic disease. The big difference between these so-called “epigenetic” changes and genetic mutations, is that the former do not result in changes in human DNA, the hallmark of a mutation, but rather alter when, and to what degree, certain genes are expressed.

Epigenetic mechanisms lead to changes in gene expression and function (Skinner, 2007). Epigenetic-driven changes in gene function occur predominantly through changes in chromatin structure, histone modification, and DNA methylation. Herbicides are among the environmental chemicals that can contribute to disease via epigenetic mechanisms.

Epigenetics play a role in the etiology of many human diseases, ranging from cancer, autoimmune disease, prostate and kidney disease, psychiatric and neurological disorders, and several others (Collotta et al., 2013; Rissman and Adli, 2014).

This important study by Cohn et al. found that breast cancer risks among female children were higher if their mother’s had been exposed to DDT.

A compelling study by Cohn et al. (2015) linked maternal exposure to DDT to heightened breast cancer risk among female children (OR=3.7), reinforcing the value of taking and storing samples of maternal blood during the first trimester of pregnancy, and infant blood at the time of birth.

Animal studies conducted by Dr. Michael Skinner’s team at Washington State University have added a troublesome new twist in the study of epigenetics. Their work has shown that exposure to certain pesticides can trigger heritable, multi-generational impacts on development, shifting the ratio of male to female offspring and increasing the frequency of some adult-onset, chronic disease (Crews et al., 2007; Manikkam et al., 2014; Nilsson et al., 2012; Skinner et al., 2013a and 2013b; Skinner, 2015).

New science has raised new concerns over epigenetic impacts from herbicide exposure that alter DNA in a way that can be passed onto future generations.

For several years, scientists thought that epigenetic changes would be isolated and impact only one generation of animals, unlike true genetic mutations that permanently alter genomes. But Skinner et al’s work proved otherwise, by showing that epigenetic changes can also be passed along from one generation to the next, despite no permanent change in an organism’s DNA. How this occurs remains a major focus of ongoing research.

Rodent studies have shown that exposure to the fungicide vinclozolin, as well as several chemical mixtures, induced transgenerational inheritance of ovarian disease (Nilsson et al., 2012).

Another rat experiment reported that exposure to the insecticide DDT increases the risk of obesity and associated diseases two or three generations later via heritable, epigenetic mechanisms (Skinner et al., 2013b). This body of work raises important scientific, regulatory, and ethical questions about health and public safety (Kabasenchle and Skinner, 2014).

 Next Section-->>>