Don't Miss Out

Subscribe to OCA's News & Alerts.

Study Links Low Birth Weight to Low Levels of Commonly Used Chemicals Found in Consumer Products


Polyfluorinated compounds (PFCs) are human made and have been used in a variety of consumer and industrial applications for a long time.  Recent studies have shown that these compounds persist in the environment, do not break down and that there is widespread exposure in wildlife and in humans (see map from EWG).  PFCs have been used in a wide variety of consumer products ranging from the teflon used for stick proof cookware, to Scotchgard used in stain repellents for furniture upholstery and carpeting.  Some insecticides and fire fighting foams also use PFCs.

Once in the body, PFCs do not break down and they are eliminated from the body very slowly.  Indeed, it takes on average 4.4 years to remove half of the PFCs in the body due to excretion.  However, given the ubiquitous and possibly daily exposure to these chemicals, humans are accumulating PFCs faster than their bodies are removing them.

PFOS and PFOA have been shown in animal studies to induce developmental and reproductive toxicity, neonatal mortality, reduced birth weight, and delays in postnatal growth.  However, these outcomes were observed using doses considerably higher than the levels humans were thought to be exposed to.  Moreover, it has been known for decades by both bird enthusiasts and veterinarians that Teflon pans that are too hot release a gas which can be deadly to birds.          

Apelberg, BJ, FR Witter, JB Herbstman, AM Calafat, RU Halden, LL Needham and LR Goldman. 2007. Cord serum concentrations of perfluorooctane sulfonate (PFOS) and perfluorooctanoate (PFOA) in relation to weight and size at birth. Environmental Health Perspectives, in press.

New data link low birth weight and body mass to very low levels of commonly used chemicals found in consumer products ranging from Teflon coated cookware to water and stain repelling textiles.

Vital statistics of newborns born at a city hospital in Baltimore, MD such as weight, length and head circumference were measured and these statistics were compared to the levels of these compounds detected in the babies' cord blood.  Babies with higher levels of these compounds tended to be slightly but significantly smaller than those with lower exposure.           

Vapors volatilizing off non-stick pans are one source of exposure to perfluorinated compounds.

The compounds in question belong to a family of chemicals called perfluorinated compounds (PFCs). Two of the chemicals most commonly detected in the environment are perfluorooctane sulfonate and perfluorooctoanoate, or PFOS and PFOA for short.

What did they do?

Apelberg et al. studied a group of babies born at the Johns Hopkins Hospital in Baltimore, MD, measuring levels of PFOS and PFOA in cord blood and analyzing how those levels were related to a series of health parameters at birth. They collected cord blood specimens from 341 babies born as singleton births (not twins), but set aside 48 of the samples, either because they didn't get enough blood (42 babies) or because the babies' medical conditions would have complicated the analysis (five with congenital deformaties and one pregnancy that began as twins). Of the 293 that were available for use in the study, 65 were born via C-section; the remainder --228-- were natural births. The blood samples were sent to the US Centers for Disease Control for chemical analysis. Details of the analyses and sampling were described in an earlier paper.

Medical records for the mothers were used to document age, race, education, body mass index of pre-pregnancy weight, smoking status, gestational age of the baby at delivery and other general information about maternal health.  The babies' birth weight in grams, head circumference and length in cm were also collected.  Serum continine levels along with medical history were used to determine maternal smoking status. An index of a baby's total mass called the ponderal index was also reported based on the birth weight in grams divided by the cube of the length in centimeters (multiplied by 100).

In their analysis they adjusted for a range of factors that can confuse the relationship between cause and effect, including smoking, maternal health status, age, race, pre-pregnancy BMI, prior preterm births, diabetes and hypertension.  These characteristics are all known to affect birth weight/gestational age and thus it was important for their analysis to incorporate them.

What did they find?

Apelberg et al. detected PFOA and PFOS in more than 99% of the samples analyzed, with the median PFOA concentration being 1.6 ng/ml, and median PFOS concentration of 5 ng/ml.  The concentration range of PFOA was 0.3 to 7.1 ng/ml, and the PFOS range was between 0.2 to 34.8 ng/ml.  PFOA and PFOS levels were highly correlated.

They found that for every 2.7 fold increase in cord blood concentration of PFOA, there is a 104 gram reduction in birth weight. For PFOS, the same increase results in a 69 gram reduction in birth weight. To put this into perspective, 70-100 grams is in the range of 2-3.25% of the mean birthweight in this population. In addition to the negative effects of PFOA or PFOS on birth weight, head circumference also decreased with increasing levels of PFOA or PFOS in natural born babies, but not C-section delivered babies.

While they found no effect of PFOA or PFOS on average length, both higher levels of PFOA and PFOS were associated with lower measurements of the baby's ponderal mass index.  The ponderal mass calculation helps to identify babies who have undergone disproportionate growth patterns (either are too tall or short for their respective weight).  Thus a decrease in ponderal mass may indicate babies who either are too fat or too thin for their respective height.

What does it mean?

Apelberg et al.'s results suggest that increasing levels of PFOA and PFOS in a baby's cord blood have a negative effect on birth weight, head circumference and body mass in this study population in Baltimore, MD.  This is the first study in humans to find this association. Additional studies in other communities nationwide will be required before these findings can be extrapolated to the population as a whole.  However, the observations made in this particular community are striking for several reasons.

The first reason is although these compounds have been around for a while, very little epidemiological data exist on the potential health effects of these compounds on humans.  While this study population is geographically limited, nearly 100% of samples tested found detectable levels of PFOA and PFOS suggesting that human exposure could be wide spread.  Given the changes observed in this population, it will be both interesting and informative to see if other patient populations possess similar sensitivity to these compounds.  This study population consisted primarily of African Americans (70%) and minorities.  It will be critical for future studies to determine if a particular demographic group is more or less exposed to PFOA and PFOS, and also if certain demographic groups are more or less sensitive to these compounds (ie, are the changes in birth weight greater or decreased in other groups).

Secondly, PFOA and PFOS have been shown to affect birth weight in rodents.  Yet the studies of rodents involved much higher doses, levels many-thousand fold higher than what Apelberg et al. were measuring in people. These data suggest that humans may be more sensitive to the effects of PFOA and PFOS than are animals. Another untested possibility is that PFOS and PFOA cause effects at low levels that can't be predicted by high dose experiments. This has become a common observation in research with contaminants that affect hormone systems. More research will be needed to test these ideas.

Thirdly, the finding that higher levels of PFOA and PFOS negatively impact head circumference raises the possibility that there are neurodevelopmental outcomes associated with PFOA and PFOS.  Head circumference is commonly used as a measure of brain growth and development, and if the findings on head circumference in this study are confirmed, it suggests the nervous system may be a critical target for PFOA and PFOS.

Despite the interesting findings from this study that suggest additional research is warranted, there were several limitations that the authors correctly point out.  Specifically, the accuracy of PFOS measurements may not be as precise as an earlier study due to required modifications to the detection equipment.  Secondly, head circumference and length measurements are more subjective measurements that can introduce measurement errors into the study. On the other hand, the most likely effect of those errors would be to make it harder to detect a real association.   Finally, there is a normal variation in the endpoints measured and the small changes observed in this study may not necessarily have clinical relevance.     


Apelberg BJ, Goldman LR, Calafat AM, Herbstman JB, Kuklenyik Z, Heidler J, et al. 2007. Determinants of fetal exposure to polyfluoroalkyl compounds in Baltimore, Maryland. Environmental Science and Technolology 41(11):3891-3897.

Environmental Working Group. 2003. PFCs: Global Contaminants

Environmental Working Group. 2005. Former Dupont top expert: Company knew.

Kannan, K, S Corsolini, J Falandysz, G Fillmann, K Kumar, BG Loganathan, MA Mohd, J Olivero, WN Van, JH Yang, KM Aldoust. 2004. Perfluorooctanesulfonate and related fluorochemicals in human blood from several countries. Environmental Science and Technology 38: 4489-4495.

York, RG 2002. Oral (gavage) two-generation (one litter per generation) reproduction study of ammonium perfluorooctanoate (APFO) in rats. Report prepared for 3M, St. Paul, MN by Argus Research (Horsham, PA). Sponsor's Study No. T-6889.6., Reviewed in US EPA AR226-1092. This document is available for download from the Environmental Working Group; it is a 22MB file.