Subscriber access provided by NEW YORK STATE DEPT OF HEALTH

Article
Perfluorinated Compounds in Human Breast Milk from Several Asian
Countries, and in Infant Formula and Dairy Milk from the United States
Lin Tao, Jing Ma, Tatsuya Kunisue, E. Laurence Libelo, Shinsuke Tanabe, and Kurunthachalam Kannan
Environ. Sci. Technol., 2008, 42 (22), 8597-8602 • Publication Date (Web): 11 October 2008
Downloaded from http://pubs.acs.org on November 19, 2008

More About This Article

Additional resources and features associated with this article are available within the HTML version:

• Supporting Information
• Access to high resolution figures
• Links to articles and content related to this article
• Copyright permission to reproduce figures and/or text from this article

Environmental Science & Technology is published by the American Chemical

Society. 1155 Sixteenth Street N.W., Washington, DC 20036
 Environ. Sci. Technol. 2008, 42, 8597–8602

Perfluorinated Compounds in Human that contained concentrations close to the limit of detection.
The estimated average daily intake of PFOS by infants from seven
Breast Milk from Several Asian Asian countries, via breastfeeding, was 11.8 ( 10.6 ng/kg bw/
day; this value is 7-12 times higher than the estimated
Countries, and in Infant Formula and adult dietary intakes previously reported from Germany, Canada,
Dairy Milk from the United States and Spain. The average daily intake of PFOA by Japanese
infants was 9.6 ( 4.9 ng/kg bw/day, a value 3-10 times greater
than the estimated adult dietary intakes reported from
L I N T A O , † J I N G M A , †,‡
TATSUYA KUNISUE,†
Germany and Canada. The highest estimated daily intakes of
E. LAURENCE LIBELO,§ PFOS and PFOA by infants from seven Asian countries studied
SHINSUKE TANABE,| AND were 1-2 orders of magnitude below the tolerable daily
K U R U N T H A C H A L A M K A N N A N * ,† intake values recommended by the U.K. Food Standards Agency.
Wadsworth Center, New York State Department of Health,
and Department of Environmental Health Sciences, School of
Public Health, State University of New York at Albany, Empire
Introduction
State Plaza, P.O. Box 509, Albany, New York 12201-0509, Perfluorinated compounds (PFCs), including perfluorooc-
School of Environmental Science and Engineering, Shanghai tanesulfonate (PFOS), perfluorohexanesulfonate (PFHxS),
Jiao Tong University, Shanghai, 200240, China, Office of perfluorooctanoate (PFOA), and perfluorononanoate (PFNA),
Pollution Prevention and Toxics, United States Environmental have been found in human blood from the general popula-
Protection Agency, Mail Code 7406C, 1200 Pennsylvania tions, from developed and developing countries alike (1-7).
Avenue, Washington, D.C. 20460, and Center for Marine Mean concentrations of PFOS and PFOA reported for
Environmental Studies, Ehime University, Bunkyo-cho 2-5, American children aged 2-12 were 37.5 and 4.9 ng/mL,
Matsuyama 790-8577, Japan
respectively; these concentrations were similar to the values
reported for adults (8, 9). PFOS and PFOA have been found
Received July 7, 2008. Revised manuscript received in cord blood samples from the United States, Germany, and
September 3, 2008. Accepted September 4, 2008. Japan (10-12). Concentrations of PFOS and PFOA in cord
blood were correlated with the concentrations in maternal
blood, indicating transplacental transfer of PFCs and in utero
exposures. PFCs were also reported to be present in the blood
The occurrence of perfluorinated compounds (PFCs) in of one-day-old neonates from New York State (13). Lactational
human blood is known to be widespread; nevertheless, the transfer of PFCs has been suggested as a source of PFC
sources of exposure to humans, including infants, are not well exposure for infants based on the analysis of breast milk
understood. In this study, breast milk collected from seven from the U.S., Sweden, Germany, and China (14-17).
countries in Asia was analyzed (n ) 184) for nine PFCs, including Gestational exposure of PFOS and PFOA in laboratory
perfluorooctanesulfonate (PFOS) and perfluorooctanoate rodents has been shown to reduce gestational length, birth
(PFOA). In addition, five brands of infant formula (n ) 21) and weight, and weight gain, and to reduce postnatal survival
11 brands of dairy milk (n ) 12) collected from retail stores during early life stages (18-23). Lactational exposure reduced
the body weight gain in pups nursed by PFOS- or PFOA-
in the United States were analyzed, for comparison with PFC
exposed dams (24, 25). However, health effects of PFCs in
concentrations previously reported for breast milk from the U.S. human infants at current exposure levels, which are 3 orders
PFOS was the predominant PFC detected in almost all Asian of magnitude lower than the concentrations used in labora-
breastmilksamples,followedbyperfluorohexanesulfonate(PFHxS) tory animal exposure studies, are unknown. Concentrations
and PFOA. Median concentrations of PFOS in breast milk of PFOS and PFOA in cord blood were negatively associated
from Asian countries varied significantly; the lowest concentration with birth weight, ponderal index, and head circumference
of 39.4 pg/mL was found in India, and the highest concentration (26). Maternal PFOA concentrations were negatively associ-
of 196 pg/mL was found in Japan. The measured concentrations ated with birth weight, length, and abdominal/head cir-
were similar to or less than the concentrations previously reported cumferences in children from Denmark (27, 28). Although
from Sweden, the United States, and Germany (median, the three studies (26-28) suggested an association between
gestational PFC exposure and birth outcomes, another study
106-166 pg/mL). PFHxS was found in more than 70% of the
involving mothers occupationally exposed to PFOS did not
samples analyzed from Japan, Malaysia, Philippines, and Vietnam, find any association between maternal exposure and birth
at mean concentrations ranging from 6.45 (Malaysia) to 15.8 weight of the babies (29).
(Philippines) pg/mL. PFOA was found frequently only in samples Dietary intake is a pathway for PFC exposure in adults.
from Japan; the mean concentration for that country was Diet was reported to contribute 61% of the estimated total
77.7 pg/mL. None of the PFCs were detected in the infant- daily intake of 410 ng/person/d for PFOS and perfluoro-
formula or dairy-milk samples from the U.S. except a few samples carboxylates in adults in Canada (30). Diet was shown to
contribute 1.4 and 2.9 ng/kg bw/d for PFOS and PFOA intakes,
respectively, in German adults (31). A study on dietary intake
* Corresponding author tel: 518-474-0015; fax: 518-473-2895; of PFCs in Spain showed that daily intake of PFOS by children
e-mail: kkannan@wadsworth.org. aged 4-9 years (2.3 ng/kg bw/d) was twice as high as that
†
Wadsworth Center, New York State Department of Health, and estimated for adult males (1.07 ng/kg bw/d) (32).
Department of Environmental Health Sciences, School of Public
Breast milk is strongly recommended for newborn babies
Health, State University of New York at Albany.
‡
Shanghai Jiao Tong University. because it provides balanced nutrients and immunoglobulins
§
United States Environmental Protection Agency. to enhance the immune system and to fight against infections
|
Ehime University. (http://www.who.int/nutrition/publications/Planning_guide.
10.1021/es801875v CCC: $40.75  2008 American Chemical Society VOL. 42, NO. 22, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8597
Published on Web 10/11/2008
pdf). Recent studies on PFC measurements in breast milk have Quality Assurance and Quality Control. The analytical
focused on samples collected from North America and Europe method was evaluated by spiking known amounts of PFCs into
(14-17); little is known about PFC concentrations in breast cow milk and human milk; the average recoveries of native and
milk, or exposure levels in infants, in Asian countries. Infant labeled PFCs ranged from 78 to 100% in both of the milk matrices
formula or dairy milk is a supplemental diet given to newborn (14). The recoveries of the four 13C-labeled internal standards
babies, and is given as the sole diet when the mother is unable in this study, which were spiked into all samples prior to
to produce milk. Occurrence of PFCs in infant formula or dairy extraction, were 104 ( 17% (mean ( SD) for 13C4-PFOS, 139
milk has not been determined. ( 34% for 13C4-PFOA, 113 ( 27% for 13C2-PFNA, and 124 (
In the present study, breast milk collected from seven 44% for 13C2-PFDA in breast milk samples and in infant formula
Asian countriessJapan, India, Malaysia, Philippines, Indo- and dairy milk samples they were 84 ( 10% for 13C4-PFOS,
nesia, Vietnam, and Cambodiaswas analyzed to determine 138 ( 37% for 13C4-PFOA, 129 ( 77% for 13C2-PFNA, and
concentrations of nine PFCs: perfluorobutanesulfonate 116 ( 69% for 13C2-PFDA. Matrix spikes were prepared by
(PFBS), PFHxS, PFOS, perfluoroheptanoate (PFHpA), PFOA, spiking known amounts of 9 native PFC standards (1 ng of
PFNA, perfluorodecanoate (PFDA), perfluoroundecanoate each) into 5 randomly selected breast milk samples and
(PFUnDA), and perfluorododecanoate (PFDoDA). In addi- passed through the entire analytical procedure. The matrix
tion, infant formula from five manufacturers representing spike recoveries for PFOS, PFOA, and PFHxS were 77 ( 6%,
the majority of the U.S. market and 11 brands of dairy milk 106 ( 15%, and 91 ( 18%, respectively. The average matrix-
purchased from retail stores in Washington, DC, and Albany, spike recoveries for the other six PFCs ranged from 70% for
NY in 2007 and 2008 were analyzed. Daily intakes of PFCs PFUnDA to 94% for PFDA. Method blanks (n ) 6) and field
by infants were estimated based on the ingestion rates of blanks (n ) 2) contained trace levels of a few of the PFCs,
and PFC concentrations in breast milk (Asia), and in infant at concentrations ranging from 0.5 ( 0.5 pg/mL for PFBS to
formula and dairy milk (United States). 21.2 ( 5.4 pg/mL for PFOA. The limit of quantitation (LOQ)
was determined to be three times the average concentration
Materials and Methods found in the blanks.
Samples. A total of 184 human breast milk samples were Data Analysis. Quantification was performed using a 1/x
obtained from nursing mothers from Japan (n ) 24), India (n weighted eight-point calibration curve prepared by means of
) 39), Malaysia (n ) 13), Philippines (n ) 24), Indonesia (n ) external standards in methanol. Concentrations in milk and
20), Vietnam (n ) 40), and Cambodia (n ) 24). Samples were infant formula were subtracted from the average blank values.
primarily from women living in the major cities in each of the Reported concentrations for PFOS, PFOA, PFNA, and PFDA
countries, and were collected during 1999 to 2005. Mothers’ were corrected by respective individual internal standard
ages ranged from 17 to 40 years, with an average of 27 years. recoveries, while the concentrations for other PFCs were not
Samples from Japan were all from mothers who were nursing corrected for the recoveries of internal standards. Concentra-
for the first time. Samples from other Asian countries included tions below the LOQ were assigned 1/2 the value of the LOQ for
first-time nursing mothers and mothers who had nursed statistical analysis. Mean and median values were calculated
previous infants. Further details of the samples and sampling only if more than 70% of the samples had concentrations above
locations are given in the Supporting Information (Table S1). the LOQ. Because the data were not normally distributed,
Samples were collected through the international collaboration nonparametric statistical tests were applied to assess the
of scientists in Asian countries and were archived at -20 °C at statistical significance. The Mann-Whitney U test was used to
the Environmental Specimen Bank (ES-Bank) at Ehime Uni- compare differences in PFC concentrations among the Asian
versity, Matsuyama, Japan. A small portion of each archived countries. Spearman’s correlation analysis was used to examine
milk sample was analyzed for PFCs at the Wadsworth Center, the relationship between mother’s age and PFC concentrations.
New York State Department of Health. Statistical significance was set at the level of p e 0.05.
In addition to the collection of breast milk samples, infant
formula samples from five manufacturers representing >99% Results and Discussion
of the U.S. market (n ) 21) were purchased from retail stores PFC Concentrations in Breast Milk, Infant Formula, and
in Washington, DC and Boston, MA, in 2007. Most of the infant Dairy Milk. Concentrations of PFCs in breast milk samples
formula samples were milk- or soy-based, purchased as powders collected from the three cities in India did not differ significantly
and ready to use or concentrated liquids, packaged in plastic, from one another and therefore were treated as one group.
glass, cardboard, or metal containers. Both organic and non- Samples from the Philippines, collected in 2000 and in 2004,
organic formulations were included. Further details of the infant were also grouped together, since no temporal difference in
formula samples are given in the Supporting Information (Table PFC concentrations was found between the two sampling
S2). Eleven brands of dairy milk were purchased (n ) 12) from periods. PFOS was the predominant compound found in the
retail stores and local farms in Albany, NY in 2008. Dairy milk Asian human breast milk samples; PFOS was detected in 85%
samples were primarily homogenized cow milk, packaged in of the samples from India and in 100% of the samples from the
plastic or paper containers. Infant-formula and dairy-milk other sampling countries (Table 1). The mean concentration
samples were kept at 4 °C and were opened just before analysis, of PFOS in breast milk ranged from 46.1 pg/mL in samples
prepared as recommended. For example, powdered formula from India to 232 pg/mL in samples from Japan. PFHxS was
samples were dissolved in water (Milli-Q water was used), and the second predominant PFC, detected overall in more than
the concentrated liquid formula samples were diluted 1:1 with 50% of the breast milk samples. PFOA was found in 96% of the
Milli-Q-water before analysis. Ready-to-feed liquid formulas samples from Japan at a mean concentration of 77.7 pg/mL.
and dairy milk were analyzed directly from the package. PFBS, PFHpA, and PFNA were found in very few of the samples
Analysis. All of the samples were extracted with Oasis analyzed. Other long-chain perfluorinated carboxylates such
weak anion exchange cartridges (WAX; 6 cc, 150 mg; Waters, as PFDA, PFUnDA, and PFDoDA were not detected in any
Milford, MA) following the solid-phase extraction (SPE) sample at concentrations above the LOQ.
method described previously (14). Target compounds were PFC concentrations in infant formula from the United States
identified and quantified by high-performance liquid chro- were 10-fold lower than the concentrations found in the Asian
matograph (HPLC, Agilent 1100 Series) coupled with an API breast milk samples. PFOS (LOQ ) 11.0 pg/mL) was detected
2000 electrospray triple quadrupole mass spectrometer (ESI- in one formula sample at a concentration of 11.3 pg/mL. PFHxS
MS/MS, Applied Biosystems; Foster City, CA). Further details (LOQ ) 1.35 pg/mL) was detected in two formula samples at
of the analysis are given in Supporting Information. concentrations of 1.36 and 3.59 pg/mL. Other PFCs were not

8598 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 22, 2008
TABLE 1. Summary of PFCs Concentrations (pg/mL) in Breast Milk Samples from Seven Asian Countries and in Infant-Formula and
Dairy-Milk Samples from the United Statesa
PFOS PFOA PFHxS PFNA PFBS PFHpA
Breast milk
Japan (Ehime; 1999), n ) 24
range 140-523 <42.5-170 <1.66-18.2 <8.82-23.9 <1.11 <4.45-13.9
positive (%) 100 92 92 13 0 25
mean (median) 232 (196) 77.7 (67.3) 7.55 (6.45)

Malaysia (Penang; 2003), n ) 13

range 48.7-350 <42.5-90.4 <1.66-13.3 <8.82-14.9 <1.11-16.5 <4.45
positive (%) 100 23 85 8 8 0
mean (median) 121 (111) 6.45 (6.68)

Philippines (Quezon; 2000 and 2004), n ) 24

range 27.0-208 <42.5-183 <1.66-58.9 <8.82-25.0 <1.11-17.4 <4.45-34.8
positive (%) 100 29 92 17 13 8
mean (median) 97.7 (104) 15.8 (13.3)

Indonesia (Jakarta, Purwakarta; 2001), n ) 20

range 25.4-256 <42.5 <1.66-6.23 <8.82-135 <1.11 <4.45-7.32
positive (%) 100 0 45 5 0 5
mean (median) 83.6 (67.2)

Vietnam (Hanoi, Ho Chi Minh; 2000 and 2001), n ) 40

range 16.9-393 <42.5-89.2 <1.66-26.8 <8.82-10.9 <1.11-3.98 <4.45-6.88
positive (%) 100 3 73 5 10 18
mean (median) 75.8 (58.5) 6.81 (4.33)

Cambodia (Phnom Penh; 2000), n ) 24

range 17.2-327 <42.5-132 <1.66-18.6 <8.82-12.3 <1.11-4.86 <4.45-38.4
positive (%) 100 4 13 13 8 29
mean (median) 67.3 (39.9)

India (Chidambaram, Kolkata, Chennai; 2002, 2004, and 2005), n ) 39

range <11.0-120 <42.5-335 <1.66-13.3 <8.82 <1.11-8.86 <4.45
positive (%) 85 8 36 0 5 0
mean (median) 46.1 (39.4)
Infant formula
United States (Washington, DC; 2007), n ) 21
range <11.0-11.3 <48.3 <1.35-3.59 <2.20 <0.7 <1.2
positive (%) 5 0 10 0 0 0
Dairy milk
United States (Albany, NY; 2008), n ) 12
range <11.0 <48.3 <1.35-3.82 <2.20 <0.7 <1.2
positive (%) 0 0 8 0 0 0
a
Mean and median values were calculated when more than 70% samples were positively detected in concentrations
above the quantitation limit (LOQ). PFDA, PFUnDA, and PFDoDA were not found in any of the samples analyzed at
concentrations above the LOQs (17.8, 24.0, and 12.7 pg/mL).

detected in any of the formula samples analyzed. Concentra- countries, in the present study, were within an order of
tions of PFCs in dairy-milk samples were similar to values found magnitude of the concentrations reported for the United States
for the infant formula samples. None of the PFCs was detected and Europe (Figure S1). The median concentration of PFOS
in any of the dairy milk samples. PFHxS was found in one soy- reported in breast milk was higher in Hungary (330 pg/mL)
milk sample, at a concentration of 3.82 pg/mL. than in the other countries studied so far (16). Concentrations
Geographical Distribution. Among the seven Asian coun- of PFOS in breast milk from Japan, in the present study, were
tries studied, the highest concentrations of PFOS and PFOA similar to the concentrations reported in Sweden (median: 166
were found in breast milk samples from Japan. Concentrations pg/mL) (14). Concentrations of PFOS in breast milk from
of PFOS and PFOA were significantly higher in breast milk Malaysia and the Philippines were similar to the concentrations
collected from Japan than from the other Asian countries (p < reported for the United States (median: 106 pg/mL) (13),
0.05) (Figure 1; Table 1). Concentrations of PFOS in breast milk Germany (median: 123 pg/mL) (16), and China (median: 100
collected from Malaysia, Philippines, and Indonesia were pg/mL) (15). Concentrations of PFCs in breast milk from
significantly higher than the concentrations in samples from Vietnam, Cambodia, and India were 40-50% lower than the
Vietnam, Cambodia, and India. The lowest concentration of concentrations reported for the United States, Germany, and
PFOS was found in breast milk from India. The concentrations China.
of PFHxS were significantly higher in milk from the Philippines The geographical pattern of PFOS distribution in breast milk
than in milk from Japan, Vietnam, and Malaysia (Figure 2); for from Asian countries resembled the pattern reported for PFOS
the remaining three countries, PFHxS was infrequently detected. in blood from those countries (3). For example, concentrations
PFOS and other PFCs have been reported in breast milk of PFOS were reported to be the lowest in human blood from
from China, Sweden, Germany, Hungary, and the United States India (3), a pattern similar to that in breast milk, in this study.
(14-17). PFOS concentrations in breast milk from Asian Mean serum concentrations of PFOS in female donors from

VOL. 42, NO. 22, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8599
 The breast milk samples analyzed in this study had been
previously analyzed for persistent organic pollutants such as
polychlorinated biphenyls (PCBs) and polybrominated diphenyl
ethers (PBDEs) (37, 38). Concentrations of PBDEs in breast milk
from Japan, Malaysia, and the Philippines were greater than
the concentrations reported for Cambodia, Vietnam, and India
(37). That finding is similar to what we have found for PFOS,
suggesting some similarities in human exposures to PFOS and
PBDEs exist in these Asian countries (Figure S2). PFOS has been
used as a surface protectant in many consumer products such
as textiles, leathers, and food packaging, to repel oil, dust, and
grease (33). PBDEs have been used as flame retardants in various
consumer products such as textiles, plastics, and electronic
equipment, to impart resistance to flammability (37). It is
believed that consumer products treated with PFCs and/or
PBDEs are potential sources of exposure to PFOS and PBDEs
in the general population.
Nursing History and PFC Concentrations. Breast milk
samples analyzed in this study, except for those from Japan,
were collected from mothers who had nursed previously as
well as those who were nursing for the first time. Previous
nursing can potentially reduce the concentrations of PFOA in
FIGURE 1. Mean concentration (pg/mL) of PFOS (gray bars) and breast milk (14). However, in this study, the concentrations of
total PFC (sum of PFOS, PFOA, PFHxS, PFNA, PFBS, and PFHpA) PFOS and PFHxS in the milk of mothers who were nursing for
(black bars) in breast milk samples (n ) 184) from seven Asian the first time were not statistically different from those of
countries.
mothers who had nursed previously. Nevertheless, median
PFOS concentrations in the milk of Malaysian and Cambodian
mothers, who were nursing for the first time were 27-100%
higher than concentrations in the milk of mothers who had
previously nursed, from the same countries (Table S3). However,
confidence in this interpretation is tempered by uncertainties
in the grouping of samples based on the number of previous
children.
Correlations among PFCs. Concentrations of PFOS, PFOA,
and PFHxS in breast milk from Japan were significantly
correlated with each other (Spearman’s coefficient, p < 0.01).
PFOS and PFHxS were significantly correlated (Spearman’s, p
< 0.05) in breast milk from the Philippines, Vietnam, and
Malaysia. Correlations between PFOS and PFHxS, as well as
between PFOS and PFOA, in breast milk from Japan suggest
FIGURE 2. Concentration of PFHxS (pg/mL) in breast milk from coexposure to several PFCs.
the Philippines (n ) 24), Japan (n ) 24), Vietnam (n ) 40), and Concentrations of PFOS, PFOA, and PFHxS were not
Malaysia (n ) 13). Asterisk indicates a statistically significant correlated with mother’s age when samples were analyzed
(p e 0.05) higher concentration in the Philippines than in the individually for each of the countries. Nevertheless, when all
other three countries. of the samples were considered together, concentrations of
PFOS in breast milk were significantly positively correlated
(Spearman’s, p < 0.001, Figure S3) with the age of the mother.
Japan, Malaysia, and India were 20, 12, and 2.3 ng/mL, This comparison (n ) 72) suggests increase in PFOS burden
respectively (3); a similar trend was found in breast milk, with with age.
PFOS concentrations of 232, 121, and 46.1 pg/mL for Japan, Daily Intake of PFCs via Breast Milk, Infant Formula,
Malaysia, and India, respectively (Figure S2). and Dairy Milk. The mean consumption level of breast milk
High concentrations of PFCs in breast milk from Japan can by infants between ages of 1 and 6 months is 742 mL/day,
be related to the production and usage of a number of consumer according to the United States Environmental Protection Agency
products that contain PFCs (33). Nevertheless, since 2001, (EPA) (39); this consumption rate is consistent with the rates
reductions in PFOS concentrations in human blood, as well as reported in several other countries (39, 40). Using this value,
in biota from the United States have been reported (34, 35), we estimated the average daily intake (ADI) of PFOS via
due to the phase out in the production of PFOS-based breastfeeding by infants with an average body weight (bw) of
compounds by 3M Company, the leading manufacturer. The 6 kg to be 11.8 ( 10.6 ng/kg bw/day, for the seven Asian countries
breast milk samples from Japan were collected in 1999, a few studied; this intake value varied from 5.7 ( 4.4 ng/kg bw/day
years earlier than the samples collected from other countries. for Indian infants to 28.7 ( 11.7 ng/kg bw/day for Japanese
There is no information available on the production of PFCs infants (Table 2). The ADI of PFOA was 9.6 ( 4.9 ng/kg bw/day
in Asian countries other than Japan and China. It is worth for Japanese infants. The ADI of total PFCs (sum of PFOS, PFOA,
noting that the concentrations of PFOS in breast milk from PFHxS, PFBS, PFHpA, and PFNA) was 18.2 ( 14.3 ng/kg bw/
countries with high gross domestic product (GDP) (e.g., day, over all of the seven Asian counties studied. Based on the
United States, Sweden, Hungary, Germany, and Japan) (36) highest concentrations measured, the highest ADIs for PFOS,
were higher than those in countries with low GDP (e.g., PFOA, and total PFC were estimated to be 64.6, 41.3, and 88.7
Vietnam, Indonesia, India, and Cambodia). In addition to ng/kg bw/day, respectively. Based on the concentrations of
economic conditions, sociological and cultural factors, as well PFCs determined in infant formula and dairy milk from the
as individual consumer habits, can contribute to the geo- United States, the maximum ADI of PFOS by infants was
graphical variability in exposure levels of PFCs. estimated to be e1.4 ng/kg bw/day; this value is an order of

8600 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 22, 2008
 samples from Vietnam, Cambodia, and India were approximately
TABLE 2. Estimated Daily Intakes (ng/kg bw/d) of PFOS and 3-5 fold lower than the concentrations found in breast milk from
Total PFCs by Infants via Breastfeeding in Asian Countries, Japan. Concentrations of PFCs in infant-formula and dairy-milk
and via Infant Formula or Dairy Milk in the United Statesa samples from the United States were approximately an order of
magnitude lower than the concentrations in breast milk. The daily
PFOS SUM1b SUM2c intake of PFOS by Asian infants via breastfeeding was 7-12 times
higher than the dietary intakes previously reported for adult
via breast milk mean SD mean SD mean SD
Japan 28.7 11.7 39.2 16.3 40.5 16.7
populations in Canada, Germany, and Spain. Further studies are
Malaysia 15.0 9.3 19.7 12.0 20.9 12.3 needed to assess the effects of PFCs in early life stages.
Philippines 12.1 5.9 19.8 10.7 21.2 11.4
Indonesia 10.3 7.0 13.2 7.0 14.9 10.0 Acknowledgments
Vietnam 9.4 8.7 13.1 9.3 14.1 9.3 We thank the following scientists for help with collection of the
Cambodia 8.3 9.4 11.7 10.1 13.1 10.4
India 5.7 4.4 10.1 7.9 11.0 8.1 breast milk samples: Dr. Yuko Kuroda from Ehime Prefectural
average 11.8 10.6 17.0 13.8 18.2 14.3 University of Health Sciences, Japan for Japanese samples; Dr.
Pham Hung Viet from Hanoi National University, Vietnam and
maxd max max Dr. Bui Cach Tuyen from University of Agriculture and Forestry,
Vietnam, for Vietnamese samples; Dr. Touch Seang Tana from
via infant formula - U.S. 1.4 4.8 5.1 Social and Culture Observation Unit of the Cabinet of the Council
via dairy milk - U.S. 0.7 4.1 4.4 of Minister, Cambodia, for Cambodian samples; Dr. Tussy A.
a
Calculation is based on an average milk intake rate of Adibroto from Agency for the Assessment and Application of
742 mL/day and an average body weight of 6 kg. b Sum of Technology, Indonesia, for Indonesian samples; Dr. Maricar
PFOS, PFOA, and PFHxS. c Sum of PFOS, PFOA, PFHxS, Prudente from De La Salle University, Philippines, for Philippine
PFBS, PFHpA, and PFNA. d One-half the LOQ value was samples; Ms. Hatijah Hashim from Consumers’ Association of
used for calculation when maximum concentration was Penang, Malaysia, for Malaysian samples; Dr. Annamalai Subra-
below the LOQ.
manianfromEhimeUniversity,JapanandMs.ParomitaChakraborty
fromTheChineseAcademyofSciences,China,forIndiansamples.
magnitude lower than the ADI previously reported from breast Financial support for this study was partly provided by Grants-
milk consumption in the United States (14). in-AidforScientificResearch(S)(20221003)fromtheJapanSociety
The calculated exposures to PFOS and other PFCs via for the Promotion of Science (JSPS) and the Waste Management
breastfeeding were at least an order of magnitude higher than Research Grants (K1821 and K1836) from the Ministry of Environ-
the exposures via formula feeding. Although formula feeding ment, Japan. This study was partly supported by the “Global COE
can reduce PFC exposures, breastfeeding provides unique Program” of the Japanese Ministry of Education, Culture, Sports,
nutrients and antibodies to strengthen infants’ immune systems. Science and Technology (MEXT), Japan and JSPS. This work was
As long as the intake of an environmental contaminant/toxicant reviewed by EPA and approved for publication but does not
via breastfeeding does not exceed the reference doses suggested necessarily reflect official Agency policy.
by national/international health organizations, breastfeeding
is still recommended by several agencies and advisory groups. Supporting Information Available
Currently, there is no consensus reference dose established for Details of analytical methods and tables showing breast milk and
the intake of PFOS and PFOA. The U.K. Food Standards Agency, infant formula samples, concentrations of PFOS and PFHxS in
Committee on Toxicology recommended tolerable daily intake relation to parity, geographical comparison of serum and breast
(TDI) values of 300 and 3000 ng/kg bw/day for PFOS and PFOA, milk PFOS and PBDE concentrations, association between age
respectively, based on a thorough review of the findings of and PFOS concentrations, and median concentration of PFOS
currently available toxicity studies (41, 42). The German Federal (pg/mL) in breast milk from seven Asian countries (black bar; this
Institute for Risk Assessment determined a TDI of 100 ng/kg study) compared with the concentrations reported for the U.S.,
bw/day for PFOS by applying an uncertainty factor of 1000 to China, Germany, Sweden, and Hungary. This material is available
a NOAEL value from a two-generation reproductive toxicity free of charge via the Internet at http://pubs.acs.org.
study in rats (43). The Environmental Working Group (EWG)
in the United States estimated a reference dose (RfD) of 25 Literature Cited
ng/kg bw/day for PFOS, based on increase in mammary
(1) Calafat, A. M.; Kuklenyik, Z.; Caudill, S. P.; Reidy, J. A.; Needham,
fibroadenomas in a chronic toxicity study in rats (44). The
L. L. Perfluorochemicals in pooled serum samples from United
estimated ADI of PFOS in Asian infants in our study did not, States residents in 2001 and 2002. Environ. Sci. Technol. 2006,
in general, exceed the TDIs recommended by the U.K. and 40, 2128–2134.
Germany, but it did exceed the RfD of EWG for certain individual (2) Olsen, G. W.; Church, T. R.; Miller, J. P.; Burris, J. M.; Hansen,
samples. K. J.; Lundberg, J. K.; Armitage, J. B.; Herron, R. M.; Medh-
The estimated ADI of PFOS by infants from the seven Asian dizadehkashi, Z.; Nobiletti, J. B.; O’Neill, E. M.; Mandel, J. H.;
Zobel, L. R. Perfluorooctanesulfonate and other fluorochemicals
countries via breastfeeding was 7-12 times higher than the in the serum of American red cross adult blood donors. Environ.
adult dietary intakes values previously reported; the average Health Perspect. 2003, 111, 1892–1901.
dietary intake of PFOS by Canadian adults (>12 years old), (3) Kannan, K.; Corsolini, S.; Falandysz, J.; Fillman, G.; Kumar, K. S.;
Germans (17-45 years old), and Spanish adults (>10 years old) Loganathan, B. G.; Mohd, M. A.; Olivero, J.; Wouwe, N. V.; Yang,
were estimated to be 1.6, 1.4, and 1.07 ng/kg bw/d, respectively J. H.; Aldous, K. M. Perfluorooctanesulfonate and related
(30-32). Our ADI values of PFOS for Asian infants were 4-6 fluorochemicals in human blood from several countries. En-
viron. Sci. Technol. 2004, 38, 4489–4495.
times higher than the average dietary intakes estimated for (4) Harada, K.; Saito, N.; Inoue, K.; Yoshinaga, T.; Watanabe, T.;
Spanish children (aged 4-9 years) (32). ADIs of PFOA by Sasaki, S.; Kamiyama, S.; Koizumi, A. The influence of time, sex
Japanese infants were 3-10 times higher than the dietary intakes and geographic factors on levels of perfluorooctanesulfonate
estimated for Canadian and German adults (1.0 and 2.9 ng/kg and perfluorooctanoate in human serum over the last 25 years.
bw/d) (30, 31). J. Occup. Health 2004, 46, 141–147.
(5) Yeung, L-W. Y.; So, M. K.; Jiang, G.; Taniyasu, S.; Yamashita, N.;
In summary, PFOS, PFHxS, and PFOA were found in breast
Song, M.; Wu, Y.; Li, J.; Giesy, J. P.; Guruge, K. K.; Lam, P-K. S.
milk samples collected from seven Asian countries, suggesting Perfluorooctanesulfonate and related fluorochemicals in human
the prevalence of PFC exposure in populations in both developing blood samples from China. Environ. Sci. Technol. 2006, 40, 715–
and developed countries. Concentrations of PFOS in breast milk 720.

VOL. 42, NO. 22, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8601
 (6) Kärrman, A.; Mueller, J. F.; van Bavel, B.; Harden, F.; Toms, (24) Luebker, D. J.; Case, M. T.; York, R. G.; Moore, J. A.; Hansen,
L-M. L.; Lindström, G. Levels of 12 perfluorinated chemicals in K. J.; Butenhoff, J. L. Two-generation reproduction and cross-
pooled Australian serum, collected 2003-2003, in relation to foster studies of perfluorooctanesulfonate (PFOS) in rats.
age, gender, and region. Environ. Sci. Technol. 2006, 40, 3742– Toxicology 2005, 215, 126–148.
3748. (25) Wolf, C. J.; Fenton, S. E.; Schmid, J. E.; Calafat, A. M.; Kuklenyik,
(7) Calafat, A. M.; Wong, L-Y.; Kuklenyik, Z.; Reidy, J. A.; Needham, Z.; Bryant, X. A.; Thibodeaux, J.; Das, K. P.; White, S. S.; Lau,
L. L. Polyfluoroalkyl chemicals in the U.S. population: Data C. S.; Abbott, B. D. , Developmental toxicity of perfluorooctanoic
from the national health and nutrition examination survey acid in the CD-1 mouse after cross-foster and restricted
(NHANES) 2003-2004 and comparisons with NHANES 1999- gestational exposures. Toxicol. Sci. 2007, 95, 462–473.
2000. Environ. Health Perspect. 2007, 115, 1596–1602. (26) Apelberg, B. J.; Witter, F. R.; Herbstman, J. B.; Calafat, A. M.;
(8) Olsen, G. W.; Church, T. R.; Hansen, K. J.; Burris, J. M.; Butenhoff, Halden, R. U.; Needham, L. L.; Goldman, L. R. Cord serum
J. L.; Mandel, J. H.; Zobel, L. R. Quantitative evaluation of concentrations of perfluorooctanesulfonate (PFOS) and per-
perfluorooctanesulfonate (PFOS) and other perfluorochemicals fluorooctanoate (PFOA) in relation to weight and size at birth.
in the serum of children. J. Child. Health 2004, 2, 53–76. Environ. Health Perspect. 2007, 115, 1670–1676.
(9) Olsen, G. W.; Huang, H. Y.; Helzlsouer, K. J.; Hansen, K. J.; (27) Fei, C.; McLaughlin, J. K.; Tarone, R. E.; Olsen, J. Perfluorinated
Butenhoff, J. L.; Mandel, J. H. Historical comparison of per- chemicals and fetal growth: a study within the Danish national
fluorooctanesulfonate, perfluorooctanoate, and other fluoro- birth cohort. Environ. Health Perspect. 2007, 115, 1677–1682.
chemicals in human blood. Environ. Health Perspect. 2005, 113, (28) Fei, C.; McLaughlin, J. K.; Tarone, R. E.; Olsen, J. Fetal growth
539–45. indicators and perfluorinated chemicals: a study in the Danish
(10) Apelberg, B. J.; Goldman, L. R.; Calafat, A. M.; Herbstman, J. B.; national birth cohort. Am. J. Epidemiol. 2008, 168, 66–72.
Kuklenyik, Z.; Heidler, J.; Needham, L.; Halden, R. U.; Witter, (29) Grice, M. M.; Alexander, B. H.; Hoffbeck, R.; Kampa, D. M. Self-
F. R. Determinants of fetal exposure to polyfluoroalkyl com- reported medical conditions in perfluorooctanesulfonyl fluoride
pounds in Baltimore, Maryland. Environ. Sci. Technol. 2007, manufacturing workers. J. Occup. Environ. Med. 2007, 49, 722–
41, 3891–3897. 729.
(30) Tittlemier, S. A.; Pepper, K.; Seymour, C.; Moisey, J.; Bronson,
(11) Inoue, K.; Okada, F.; Ito, R.; Kato, S.; Sasaki, S.; Nakajima, S.;
R.; Cao, X. L.; Dabeka, R. W. Dietary exposure of Canadians to
Uno, A.; Saijo, Y.; Sata, F.; Yoshimura, Y.; Kishi, R.; Nakazara,
perfluorinated carboxylates and perfluorooctane sulfonate via
H. Perfluorooctanesulfonate (PFOS) and related perfluorinated
consumption of meat, fish, fast foods, and food items prepared
compounds in human maternal and cord blood samples:
in their packaging. J Agric. Food Chem. 2007, 55, 3203–10.
assessment of PFOS exposure in a susceptible population during
(31) Fromme, H.; Schlummer, M; Möller, A.; Gruber, L.; Woĺz, G.;
pregnancy. Environ. Health Perspect. 2004, 112, 1204–1207.
Ungewiss, J.; Böhmer, S.; Dekant, W.; Mayer, R.; Liebl, B.;
(12) Midasch, O.; Drexler, H.; Hart, N.; Beckmann, M. W.; Angerer, Twardella, D. Exposure of an adult population to perfluorinated
J. Transplacental exposure of neonates to perfluorooctane- substances using duplicate diet portions and biomonitoring
sulfonate and perfluorooctanoate: a pilot study. Int. Arch. Occup. data. Environ. Sci. Technol. 2007, 41, 7928–7933.
Environ. Health 2007, 80, 643–648. (32) Ericson, I.; Martĺ-Cid, R.; Nadal, M.; Bavel, B. V.; Lindström, G.;
(13) Splithoff, H.; Tao, L.; Shaver, S.; Kannan, K.; Aldous, K.; Pass, Domingo, J. L. Human exposure to perfluorinated chemicals
K.; Eadon, G. The use of newborn screening program blood through the diet: intake of perfluorinated compounds in foods
spots for exposure assessment: declining levels of perfluorinated from the Catalan (Spain) market. J. Agric. Food Chem. 2008, 56,
compounds in New York State infants. Environ. Sci. Technol. 1787–1794.
2008, 42, 5361–5367. (33) Environment Directorate, OECD. PFOS, PFAS, PFOA and Related
(14) Tao, L.; Kannan, K.; Wong, C. M.; Arcaro, K. F.; Butenhoff, J. L. Chemicals: Responses to the Questionnaire; Report ENV/JM 22;
Perfluorinated compounds in human milk from Massachusetts, Organisation for Economic Cooperation and Development:
U.S.A. Environ. Sci. Technol. 2008, 42, 3096–3101. Paris, 2004.
(15) Kärrman, A.; Ericson, I.; van Bavel, B.; Darnerud, P. O.; Aune, (34) Olsen, G. W.; Mair, D. C.; Geagen, W. K.; Ellefson, M. E.;
M.; Glynn, A.; Lignell, S.; Lindström, G. Exposure of perfluori- Ehresman, D. J.; Butenhoff, J. L.; Zobel, L. R. Preliminary evidence
nated chemicals through lactation: levels of matched human of a decline in perfluorooctansulfonate (PFOS) and perfluo-
milk and serum and a temporal trend, 1996-2004, in Sweden. rooctanoate (PFOA) concentrations in American red cross blood
Environ. Health Perspect. 2007, 115, 226–30. donors. Chemosphere 2007, 68, 105–111.
(16) So, M. K.; Yamashita, N.; Taniyasu, S.; Jiang, Q.; Giesy, J. P.; Chen, K.; (35) Kannan, K.; Perrotta, E.; Thomas, N. J. Association between
Lam, P. K. Health risks in infants associated with exposure to perfluorinated compounds and pathological conditions in
perfluorinated compounds in human breast milk from Zhoushan, southern sea otters. Environ. Sci. Technol. 2006, 40, 4943–4948.
China. Environ. Sci. Technol. 2006, 40, 2924–2929. (36) CIA World Factbook; Central Intelligence Agency: Washington,
(17) Völkel, W.; Genzel-Boroviczény, O.; Demmelmair, H.; Gebauer, DC, 2003; http://www.cia.gov.
C.; Koletzko, B.; Twardella, D.; Raab, U.; Fromme, H. Perfluo- (37) Tanabe, S.; Ramu, K.; Isobe, T.; Takahashi, S. Brominated flame
rooctanesulphonate (PFOS) and perfluorooctanoic acid (PFOA) retardants in the environment of Asia-Pacific: an overview of spatial
in human breast milk: Results of a pilot study. Int. J. Hyg. Environ. and temporal trends. J. Environ. Monit. 2008, 10, 188–197.
Health. 2008, 211, 440–446. (38) Tanabe, S.; Kunisue, T. Persistent organic pollutants in human breast
(18) Luebker, D. J.; York, R. G.; Hansen, K. J.; Moore, J. A.; Butenhoff, milk from Asian countries. Environ. Pollut. 2007, 146, 400–413.
J. L. Neonatal mortality from in utero exposure to perfluorooc- (39) U.S. Environmental Protection Agency. Child-Specific Exposure
tanesulfonate (PFOS) in Sprague-Dawley rats: dose-response, Factors Handbook; EPA/600/P-00/002B; National Center for
and biochemical and pharmacokinetic parameters. Toxicology Environmental Assessment: Washington, DC, 2002.
2005, 215, 149–169. (40) NAS (National Academy of Sciences). Nutrition during Lactation;
(19) Thibodeaux, J. R.; Hanson, R. G.; Rogers, J. M.; Grey, B. E.; Barbee, National Academy Press: Washington, DC, 1991.
B. D.; Richards, J. H.; Butenhoff, J. L.; Stevenson, L. A.; Lau, C. (41) Committee on Toxicity of Chemicals in Food, Consumer Products
Exposure to perfluorooctane sulfonate during pregnancy in rat and the Environment. COT Statement on the Tolerable Daily In-
and mouse. I. Maternal and prenatal evaluations. Toxicol. Sci. take for Perfluorooctanesulfonate; 2006; http://www.food.gov.uk/
2003, 74, 369–381. multimedia/pdfs/cotstatementpfos200609.pdf.
(42) Committee on Toxicity of Chemicals in Food, Consumer Products
(20) Lau, C.; Thibodeaux, J. R.; Hanson, R. G.; Rogers, J. M.; Grey,
and the Environment. COT Statement on the Tolerable Daily In-
B. E.; Stanton, M. E.; Butenhoff, J. L.; Stevenson, L. A. Exposure
take for Perfluorooctanoic Acid; 2006; http://www.food.gov.uk/
to perfluorooctane sulfonate during pregnancy in rat and mouse
multimedia/pdfs/cotstatementpfoa200610.pdf.
II: Postnatal evaluation. Toxicol. Sci. 2003, 74, 382–392.
(43) BfR (Bundesinstitut für Risikobewertung, German Federal
(21) Lau, C.; Butenhoff, J. L.; Rogers, J. M. The developmental toxicity Institut for Risk Assessement). High Levels of Perfluorinated
of perfluoroalkyl acids and their derivatives. Toxicol. Appl. Organic Surfactants in Fish Are Likely to Be Harmful to Human
Pharmacol. 2004, 198, 231–241. Health; Statement 021/2006, 28.7; 2006; http://www.bfr.
(22) Butenhoff, J. L.; Kennedy, G. L.; Frame, S. R.; O’Connor, J. C.; York, bund.de/cms5w/sixcms/detail.php/8172.
R. G. The reproductive toxicology of ammonium perfluorooctanoate (44) Thayer, K; Houlihan, J. Perfluorinated Chemicals: Justification
(APFO) in the rats. Toxicology 2004, 196, 95–116. for Inclusion of This Chemical Class in the National Report
(23) Lau, C.; Thibodeaux, J. R.; Hanson, R. G.; Narotsky, M. G.; Rogers, on Human Exposure to Environmental Chemicals; Environ-
J. M.; Lindstrom, A. B.; Strynar, M. J. Effects of perfluorooctanoic mental Working Group: Washington, DC, 2002.
acid exposure during pregnancy in the mouse. Toxicol. Sci. 2006,
90, 510–518. ES801875V

8602 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 22, 2008