Community Exposure to PFCs in Washington County, Minnesota:
The East Metro Perfluorochemical Biomonitoring Pilot Project
Jean Johnson PhD, Adrienne Kari MPH, Carin Huset PhD, Allan Williams PhD
Perfluorochemical (PFC) contamination of groundwater from nearby
waste disposal facilities was first discovered in Washington County,
Minnesota, located east of the Minneapolis-St. Paul metropolitan area
(also referred to as “East Metro”), in 2004 by the Minnesota Pollution
Control Agency (MPCA) and the Minnesota Department of Health
(MDH). Further investigation revealed widespread contamination of
ground water that supplied drinking water to residents in the City of
Oakdale and other communities. The 3M Company began research
and development of PFCs at its Cottage Grove facility in southern
Washington County in the 1940s; commercial production started in
the early 1950s. Wastes from the production process and wastewater
treatment plant sludge were disposed of at the facility and at several
disposal sites in Washington County between 1956 and 1971
(estimated dates).
We found measurable levels of PFOA, PFOS, and PFHxS in
specimens from all 196 participants. PFBA was detected in 28% of the
population and PFBS was detected in 3%. PFPeA and PFHxA were
below the limit of detection (0.1 ng/ml) for all specimens.
This biomonitoring project was conducted by MDH to measure the
range and distribution of PFC exposure in a population sample from
each of two communities identified as likely to be exposed. The work
was conducted in accordance with 2007 Minnesota Legislation that
established a pilot program to build MDH capacity for a state
biomonitoring program.
Mean serum concentrations were significantly higher in males for
PFOS (p<0.001) and for PFHxS (p<0.004), but not for PFOA.
PFOA, PFOS, and PFHxS were found to be significantly different
(p<0.05) across age categories and increased with age. Similar
increases in PFOA and PFHxS levels were observed with increased
length of residence in the home.
The difference in mean PFOA concentrations observed between the
municipal water and private well communities was of borderline
significance (<0.06), while community differences observed for PFOS
and PFHxS were not significant.
Table 2. PFC Serum Concentrations by Gender & Drinking Water
Community
PFC
Participant Recruitment Methods
Two communities located in Washington County were selected, each
defined based on their household drinking water supply:
PFOA
1. People currently living in households served by the Oakdale
Municipal Water Supply
Total (196)
Females (108)
Males (88)
PFOS
Median
(ng/ml)
Range
(ng/ml)
16
15
17
1.6-177
1.6-152
3.0-177
15.4
14.4
16.6
Municipal water (98)
Private well water. (98)
2. People currently living in households served by private wells in Lake
Elmo and Cottage Grove with known PFOA and/or PFOS contamination.
To be eligible, individuals had to be at least 20 years of age and lived
in their current residence prior to January 1, 2005 (a date before most
remediation efforts were initiated). In August 2008, a household
survey was mailed to 500 households randomly selected from
Oakdale municipal billing records and to all 169 homes identified
from well sampling records with a private well contaminated with > 0.1
ppb of PFOA or PFOS. From a list of eligible adults identified in the
survey, individuals were randomly selected and recruited. A total of
196 adults were enrolled, 98 individuals from each community. Each
completed a brief survey and provided a blood serum specimen for
analysis. Specimen collection was completed on December 31,
2008.
Geometric
Study Population (N)
Mean (ng/ml)
17.3
13.6
21
13
2-79
1.6-177
Total
Females
Males
Municipal water
Private well water
35.9
30.5
43.9
39.3
32.9
41
35
45.5
43
35
3.2-448
3.2-151
9.1-448
3.9-166
3.2-448
PFHxS Total
Females
Males
Municipal water
Private well water
8.4
7.0
10.6
8.6
8.3
8.9
7.4
10.5
9.8
7.5
0.32-316
0.32-316
1.7-270
0.32-72
0.37-316
Table 1. Study Participant Characteristics
Participant age
Years lived in house
Years lived in Oakdale
Mean
Min
Max
53.1
17.8
20.7
25
4
3.5
85
62
62
Private Well Community (N=98)
Participant age
Years lived in house
53.3
19.8
20
4
86
60
Figures 1-3. PFC serum concentration geometric means and 95% CI by age
category, N=196 (2003-2004 U.S. Population levels)
30.0
PFOA GeoMean (ng/mL)
Municipal Water Community (N=98)
25.0


20.0

15.0

14.4
(4.2)
5.0
8.1
(3.9)

0.0
20-39
40-59
Age Category (years)
60+
PFOS GeoMean (ng/mL)
60.0

50.0
43.4
(23.2)

40.0


30.0
20.0



PFOA and PFOS levels in water were strongly correlated with levels
found in the blood (r2=0.30 and 0.31 respectively). For PFOA, age
and water concentration together accounted for approximately 43% of
the variability in PFOA levels found in blood. For PFOS, age, gender,
and water concentration together accounted for approximately 44% of
the variability in PFOS levels found in blood.
Figure 4. PFOA drinking water concentration vs. PFOA blood
levels for 98 people with contaminated private wells
1000


 
100







  

 

 





   








 










    



 


    
 







10

1
0.01
0.1
1
10
PFOA Water Concentration ppb
Figure 5. PFOS drinking water concentration vs. PFOS blood
levels for 98 people with contaminated private wells
1000



100

10




































 




 

  

 






  







1
0.01
0.1
1
10
PFOS Water Concentration ppb
Discussion


A subsequent analysis looked at how strongly PFOA and PFOS levels
in the blood are correlated with the levels of PFCs measured in
drinking water for the 98 people for whom the levels in their private
wells were known. For each person, drinking water exposure was
estimated by taking an average of the amount of PFOS and PFOA
measured in all well samples collected from the home in 2005 through
2008. Average concentrations in water ranged from 0.04-1.87ng/ml
for PFOA and 0.04-2.47 ng/ml for PFOS. The analysis also took into
account the participant’s age, gender, and years living in the home.


10.0
20.1
(3.7)
Correlation with Drinking Water Concentrations in
Private Wells
PFOA Blood Concentration ppb
Biomonitoring Results
PFOS Blood Concentration ppb
Background and Purpose
34.7
(22.0)

Individuals living in the East Metro area prior to Jan. 1, 2005 were
exposed to PFCs through their drinking water. Serum levels were
elevated in comparison with results reported for the US general
population; they were 2 to 4 times that reported in NHANES samples
collected in 2003-2004, but comparable to or lower than levels found
in other studies of communities exposed via drinking water. Past well
water levels were strongly correlated but only partially explained
variability in the blood serum levels, indicating that other sources
likely contribute to exposure in this population.
Based on these findings, a follow-up study is recommended to
measure changes in blood levels over time. Levels are expected to
decrease due to the remedial actions taken since 2005 to remove
PFCs from the drinking water in these communities.
21.9
(18.7)

10.0
0.0
20-39
40-59
Age Category (years)
60+
Acknowledgements
Laboratory Analysis Methods
16.0

14.0
PFHxS GeoMean (ng/mL)
Serum samples were analyzed for seven PFCs by the Public Health
Laboratory at MDH using solid phase extraction and liquid
chromatography tandem mass spectrometry with stable isotope
labeled internal standards used for quantitation. One ml of serum was
extracted using Oasis HLB cartridges with a modified wash step. We
prepared matrix matched calibration curves from bovine calf serum
daily for quantification. Separations were performed using an Agilent
1200 HPLC and a water/acetonitrile mobile phase (modified with 1%
formic acid). An Applied Biosystems 4000 Qtrap tandem mass
spectrometer operated in negative ESI mode was used for detection,
with two MRM transitions monitored per analyte (when possible). The
precision and accuracy of this method, as determined through spike
and recovery experiments, are 1-8% and 99-115% respectively.
12.0

10.0


8.0

6.0

4.0



7.6
(1.9)
4.2
(1.8)
2.0
0.0
20-39
40-59
Age Category (years)
60+
12.0
(2.0)
We gratefully acknowledge the contributions to this project of
numerous MDH staff and the Environmental Health Tracking and
Biomonitoring Advisory Panel and Workgroup. We also thank the East
Metro area community members, elected representatives, medical
professionals and local public health officials for their support and
cooperation. The East Metro PFC Biomonitoring Pilot Project was
funded by the Minnesota State Legislature as part of the
Environmental Health Tracking and Biomonitoring Program at the
Minnesota Department of Health. For more information, contact the
MDH Environmental Public Health Tracking and Biomonitoring
program: www.health.state.mn.us/tracking,
[email protected], 651-201-5900 (phone).