Do OELs Really Protect Workers from Reproductive and Developmental Effects?
The latest data from the United States Census Bureau indicate that 66 percent of women who gave birth to their first child between 2006 and 2008 worked during their pregnancy. That figure is a dramatic increase from earlier generations; for example, only 44 percent of women who had their first child during the 1960s worked during pregnancy. Given this trend, the potential impact of workplace chemical exposures on reproduction and development is a significant concern. This concern involves not only exposures of pregnant women but also of male and female workers prior to conception. Furthermore, because children may be exposed to chemicals via breast milk, the period of concern for developmental toxicity does not cease at birth. Here we address the question of whether occupational exposure limits (OELs) protect workers against developmental and reproductive toxicity (DART) effects. We also identify several difficulties that industrial hygienists may face in assessing the appropriateness of an OEL for workers who are pregnant or who expect to conceive.
The basic process for developing an OEL, as described in a December 2015 supplement to the
Journal of Occupational and Environmental Hygiene
(JOEH), is to review and evaluate relevant scientific literature; select the critical health endpoint (that is, the most sensitive effect); select the point-of-departure (POD) from the key study (the dose or concentration near the low end of the observed range of the critical health endpoint); apply assessment factors that most appropriately represent the uncertainty and variability associated with the POD; and calculate the OEL as the POD divided by the assessment factors. (The calculation may involve other factors such as route-to-route extrapolation, animal-to-human conversions, or other adjustments.) Although definitions vary slightly, the resulting OEL is generally intended to protect all or the vast majority of workers against adverse effects during their working lifetime and beyond.
Numerous studies have indicated that peak concentration in maternal blood (Cmax) is more important for determining developmental effects than averaged exposure, likely because a short high-level exposure may overwhelm maternal metabolism.
Much of the data used in assessing DART concerns for workers come from toxicity tests conducted in laboratory animals (usually rats and mice) rather than studies in humans. Although inhalation is the major route of entry for hazardous chemicals in the work environment, many animal studies are conducted using oral dosing. As a result, route-to-route extrapolation is a common practice in OEL derivation. Studies in other species (fish, insects, birds) may also be available for particular compounds; however, these studies are generally not used to establish safe human exposure levels due to the greater degree of inter-species differences and the lack of standardized testing protocols. While data from studies of exposed workers are obviously highly relevant to the occupational setting, these studies are typically limited in number and are often complicated by factors such as simultaneous exposures to multiple chemicals. Understanding of DART risks for any given chemical generally requires multiple animal studies, which can be expensive and time-consuming. For these reasons, adequate DART data are lacking for most chemicals. The biology of human reproductive and developmental processes is complex.
In utero
development involves rapid division and differentiation to transform a single stem cell into the billions of highly specialized cells of the newborn infant. This process is under very tight biological control, with the timing of cell division, migration, and specialization governed by a specific set of genes. These genes are turned on and off during pregnancy, directing the development of different fetal tissues. As a result, there are well-recognized “windows of susceptibility” where fetal development may be disrupted by chemical exposure. For example, formation of many fetal organs occurs between the third and eighth week of pregnancy. It is thus important to encourage employees to inform health and safety personnel of their pregnancy (or their intended pregnancy) as early as possible, or, better, to maintain exposures for women and men below appropriately considered OELs at all times.
Historically, the scientific rationales for setting OELs were limited and aimed at minimizing or preventing significant acute toxicity. Thus, most early OELs did not include consideration of potential DART effects. Today, DART studies (if available) are commonly considered when selecting the critical effect from which to base derivation of an OEL. Agents that cause DART effects in the absence of other types of systemic health effects are considered to be selectively toxic to reproduction and development. Conversely, agents that cause DART effects only at or above doses or concentrations that cause other systemic effects in the adult may be doing so indirectly (that is, the DART effects may be secondary to parental effects). Thus, when DART effects are appropriately considered, the resulting OEL will be based either on protection of a DART effect or on a more or equally sensitive non-DART effect (which will also be protective of DART effects). A third scenario, one that indirectly considers DART effects, employs an assessment factor for database uncertainty during development of an OEL. Used when sufficient data to assess the potential for DART (or other) effects are lacking, this process results in lowering the OEL to account for missing data that might have identified a DART effect as a more sensitive endpoint.
The level of documentation and transparency in setting OELs varies widely by organization. Many of the leading organizations provide concise documentation of the underlying studies and rationale for their OELs, which often includes a section on DART effects. The basis for some ACGIH Threshold Limit Values (TLVs) is exclusively, or in part, DART effects (for example, 2-ethoxyethanol, methyl isopropyl ketone, toluene). Many of the German Maximale Arbeitsplatz-Konzentrations (MAK) recommendations explicitly address the potential for DART effects via a “MAK Pregnancy Risk Group Classification” (see Table 1). MAKs assigned to Group A or B are clearly not intended to be protective of DART effects, but those in Group C are. We have found that documentation of the consideration of DART effects in setting OELs, including the uncertainty therein, is not as robust as that of some environmental exposure criteria, such as reference concentrations (RfCs) established by EPA and Minimal Risk Levels (MRLs) established by ATSDR. The overall trend, however, in setting OELs is toward increased documentation. This trend will better serve stakeholders in assessing whether OELs are protective of DART effects.
Table 1.
MAK Pregnancy Risk Group Classification
A developing organism is most susceptible to chemical damage during fairly narrow timeframes. Sufficient exposure during one of these time periods may result in permanent effects being propagated into later stages of development. For example, as described in a 1994 article in the journal
, animal studies of the solvent 2-methoxyethanol have shown that exposures on just two days of gestation in the rat resulted in brain malformations but exposures two days later resulted in paw deformities. A related question relates to the use of an appropriate dose estimate; for most OELs, a time-weighted average (TWA) exposure, determined over eight hours, serves as the exposure metric. Ceiling levels and STELs are fairly uncommon and are typically tied to effects other than developmental toxicity. However,
numerous studies have indicated that peak concentration in maternal blood (Cmax) is more important for determining developmental effects than averaged exposure, likely because a short high-level exposure may overwhelm maternal metabolism and allow fetal exposures in a way that lower-level exposures may not. Following this logic, in theory every chemical of developmental concern should have a ceiling or short-term limit to maintain maternal Cmax within acceptable limits. However, human and animal toxicity studies typically involve steady-state exposures, in part because a study involving variable peak doses at multiple times during pregnancy to determine the worst possible exposure pattern would be incredibly complicated. As a result, the data to support such values are generally lacking. Even so, for processes involving recognized DART agents that are likely to generate highly variable exposures, it would be reasonable to limit excursions to the extent possible. Regulatory agencies are also increasingly becoming concerned with this issue; EPA recently noted that a single acute exposure to trichloroethylene during the critical developmental window can induce fetal heart defects.
When multiple organizations have published an OEL, the derived value often varies considerably. This variability may reflect differences in risk policy between organizations, such as whether the OEL takes into account technical and economic feasibility or is purely health-based; or differences in risk assessment methodology that determine, for example, how a POD is selected. Some organizations may limit the selection of critical studies to the published scientific literature, while others may allow the use of unpublished industrial research or reports. As discussed in the JOEH special issue on OELs, differences may also stem from the sometimes subjective or arbitrary selection of values used to represent assessment factors. Ultimately, the most appropriate OEL will be the one that best aligns with the target population, which in this case is pregnant women or other workers who intend to conceive. Some OELs are not intended to be protective of DART effects (for example, MAK Groups A and B) and are clearly not appropriate for this target population. OELs that are not documented to have considered DART effects should be viewed with suspicion. The most credible choice of OEL for this target population is one that is based on a DART effect, clearly considers up-to-date DART effects studies or reviews, or compensates for a lack of DART effects data via assessment factors. Unfortunately, such a choice may not exist for many chemicals. If no existing OELs account for potential DART effects of a particular chemical, or if there is no OEL at all, a number of resources might provide additional information that can be used to set an OEL. TOXNET, a group of databases managed by the National Library of Medicine, is one example. Among these databases are the Developmental and Reproductive Toxicology Database and the Hazardous Substances Data Bank, both of which contain summaries of and references to DART literature. Another source is the European Chemicals Agency (ECHA), which provides access to information, including published and unpublished DART studies, on the chemicals manufactured and imported in Europe. Interpretation of studies and data from these and other sources should be done in consultation with a scientist familiar with toxicity testing protocols and interpretation of DART data.
Tap on the image to open a larger version in your web browser.
Table 2.
Strategies for Chemicals with Poor or No DART Data
Other options for chemicals with poor or no DART data include strategies that utilize information from other chemicals thought to be similar. These strategies include Hazard Banding, Read Across, and Structural Activity Relationship (SAR) programs (see Table 2). High-throughput assays, studies with isolated cells or DNA shown to be predictive of effects in animals, are an increasingly common strategy for generating data quickly and inexpensively. The investments of time and money necessary for animal testing are stimulating efforts to develop high-throughput studies for DART effects. Due to the significant uncertainties associated with all of these strategies, they should be applied in a precautionary manner and interpreted by scientists who are familiar with these approaches.
Many workers are understandably concerned about the potential for workplace chemicals to adversely affect their child during pregnancy or their ability to conceive. To address these concerns, it is not enough to simply check that all exposures are below available OELs. Industrial hygienists should carefully review the basis of existing OELs or, when OELs are completely lacking, take advantage of available tools and expert advice to establish new ones. Routine evaluation of potential DART risks should be part of every health and safety program.
is a toxicologist and principal at Gradient in Seattle, Wash. He can be reached at
or (206) 267-2924.
is a toxicologist and industrial hygienist at Gradient in Bend, Ore. He can be reached at
or (541) 647-6066.

Readers are encouraged to review the December 2015 supplement of the
Journal of Occupational and Environmental Hygiene
, a special issue on OELs. Other sources of information in this article include:
2015 Guide to Occupational Exposure Values
Deutsche Forschungsgemeinschaft (DFG):
List of MAK and BAT Values 2015
(September 2015).
EPA: “TSCA Work Plan Chemical Risk Assessment: Trichloroethylene: Degreasing, Spot Cleaning and Arts & Crafts Uses (CASRN 79-01-6)” (
Lewandowski, TA: “Developmental Toxicology” in
Hamilton and Hardy’s Industrial Toxicology
(sixth edition, 2015).
Permissible Exposure Limits – Annotated Tables
Teratology: “Developmental phase alters dosimetry-teratogenicity relationship for 2-methoxyethanol in CD-1 mice” (March 1994).
United States Census Bureau: “Maternity Leave and Employment Patterns of First-Time Mothers: 1961-2008” (
, October 2011).
Tap on the image to open a larger version in your web browser.