Exposure Investigation at Coronet Industries
(a/k/a BORDEN FEED PHOSPHATE COMPLEX)
PLANT CITY, HILLSBOROUGH COUNTY, FLORIDA
Coronet Industries is located at 4082 Coronet Road, south of the city limits of Plant City, Florida.The company processes phosphates for use in animal food supplements, primarily for the poultryindustry. The facility has been operating for 91 years. In the past, the company mined phosphatedeposits in the area, but no mining operations are currently being conducted.
Groundwater collected from monitoring wells at the facility is contaminated with fluoride, arsenic,cadmium, lead, and alpha radiation (1). Many residents who live near the site rely on private wellsfor potable water. Public health officials and residents expressed concern that water from these wellscould contain chemical contaminants at concentrations of health concern. This contamination couldoriginate from naturally-occurring minerals in the underlying phosphate deposits or from chemicalsreleased during operations or waste management practices at the facility.
During the first two weeks of August 2003, state officials collected water samples from the privatewells of residents who lived within a ¼-mile radius of the facility. The Department of HealthLaboratories tested water samples from 43 homes for volatile organic chemicals, metals (includinglead, cadmium, arsenic, and boron), fluoride, and gross alpha and gross beta radiation. The resultsindicated that water from some of the wells contained boron, arsenic, and alpha radiation atconcentrations in excess of state drinking water standards. None of the water samples contained leador cadmium at concentrations in excess of state drinking water standards.
In order to better characterize human exposure to site-related chemicals for residents who wereconsuming the water, the Florida Department of Health (FDOH) and the Hillsborough CountyHealth Department (HCHD) assisted the Agency for Toxic Substances and Disease Registry(ATSDR) in conducting an Exposure Investigation (EI). We collected urine samples from residentswhose wells had been sampled by the state and tested the urine samples for lead, cadmium, uranium, arsenic, fluoride, and boron.
The participants in this EI met the following criteria: (1) they lived within a ¼-mile of the facility,(2) they were > 2 years old and were capable of providing a clean-catch urine sample, and (3) their well had been tested by the State.
During the week of August 12, FDOH sent potential participants a flier that alerted them to theupcoming EI. The flier informed them that urine samples would be collected the following weekfrom eligible residents. The residents were advised not to eat any fish or shellfish for four days priorto donating a urine sample. These instructions were given because after eating fish, the urinaryarsenic level could be temporarily elevated because of the high content of organic and inorganicarsenic species in fish and shellfish. Nevertheless, people who had recently eaten fish were notexcluded from the investigation, since biomarkers of exposure for the other contaminants would notbe affected.
On August 20, staff from the FDOH, HCHD, or ATSDR visited each home whose well had beentested and invited the residents to participate in the EI. Each adult participant completed a writteninformed consent form. Parents or guardians of children gave written permission to test their eligiblechildren or wards. Children, aged 7-17 years old, gave their written assent.
A 70-ml plastic urine cup was given to each participant and he/she was instructed to provide a firstmorning void sample. Written instructions were given to parents of young children to instruct themin how to assist their children in collecting a urine sample. The participants were told to place thesamples in zip lock bags containing adsorbent pads and store them in a refrigerator until they werecollected by a representative of ATSDR.
The following day, ATSDR staff collected the samples and poured an aliquot of about 10 ml of theurine into a 20-ml screw cap, plastic collection bottle. The samples were frozen on dry ice until theywere shipped to National Medical Services (NMS) in Willow Grove, Pennsylvania for fluorideanalysis. NMS measured the fluoride concentration in the urine samples using an ion-specificelectrode.
The remaining urine in the urine cups was stored frozen on dry ice until ATSDR delivered them tothe National Center for Environmental Health Laboratory (NCEH) in Atlanta for analysis. TheNCEH analyzed the samples for lead, cadmium, uranium, and arsenic using inductively coupledplasma-mass spectroscopy (ICP-MS). Creatinine concentrations were measured using an automatedspectrophotometric technique.
At the time that the protocol for this EI was developed, the results of the private well testing werenot yet available. The State later reported that water from eight of the wells they tested had a boronconcentration in excess of their drinking water advisory level. Therefore, we later offered to test theurine samples that had already been collected for boron. Of the original 106 participants, 101requested that their urine samples also be tested for boron. These urine samples were tested forboron by National Medical Services using inductively coupled plasma-atomic emissionsspectroscopy (ICP-AES).
106 residents of 35 homes completed a consent/assent form and provided a urine sample. The testpopulation consisted of 78 adults and 28 children, aged 3 to 17 years old. It was not possible toanalyze three of the urine samples for all chemicals because of the small volume of urine that wasprovided.
Data summaries for the concentrations of lead, cadmium, and uranium that were detected in the urinesamples are presented in Tables 1-3. For statistical analyses, chemical concentrations below theanalytical detection level were assigned a value equal to the detection level divided by the squareroot of 2.
The individual test results for these chemicals were compared to the 95th percentile value for thecivilian United States population as reported in the Center for Disease Control's (CDC) SecondNational Report on Human Exposure to Environmental Chemicals (2). The comparisons were madeusing creatinine-normalized values to correct for urinary dilution. Since the percentile values mayvary by age, individual comparisons were made using an age specific percentile value. For adults,gender specific comparison values were used. Most of the individual urine test results were withinthe 95th percentile of the national population, but a few individual test results did exceed the 95thpercentile. This finding is expected since, in a normal population, about 5 percent of samples wouldexceed the 95th percentile of the national comparison population.
A statistical summary of the concentrations of lead that were detected in urine samples in this EI ispresented in Table 1. None of the well water samples tested by the state laboratory had a lead concentration in excess of the state drinking water standard of 15 µg/L.
|range||ND - 4.04||ND - 3.78|
ND = not detected
Most of the EI participants had urine lead concentrations within the national reference range (95th percentile = 2.37 µg/g creatinine). However, the urine concentration of lead was elevated above the 95th percentile comparison value in three participants. There is no health based standard for urine lead concentrations. Only a few studies have simultaneously measured blood lead and urine lead concentrations in populations with low-level exposure (3, 4). In a recent study, Gulson et al. examined the correlation between blood lead and urine lead concentrations (3). This study indicated that there is a statistically significant correlation between urine and blood lead levels, although there is considerable scatter in the data, particularly at blood lead levels below 10 µg/dL. These findings suggest that the blood lead levels in the three people in this EI with an elevated urine lead level would be below 10 µg/dL. The CDC recommends that blood lead levels in children, less than 6 years old, should be below 10 µg/dL. In adults with occupational exposure to lead, the Occupational Safety and Health Administration recommends that blood lead levels should be below 40 µg/dL. Therefore, the estimated blood lead levels in the participants of this investigation would not be expected to exceed a level of health concern.
A statistical summary of the concentrations of cadmium that were detected in the urine samples inthis EI is presented in Table 2. None of the well water samples tested by the state laboratory had acadmium concentration in excess of the state drinking water standard of 5 µg/L.
|range||ND - 1.58||ND - 1.76|
ND = not detected
Most of the EI participants had urine cadmium concentrations within the national reference range (95th percentile = 1.03 µg/g creatinine). However, the urine concentration of cadmium was elevatedabove the 95th percentile comparison value in five participants. For workers with occupationalexposure to cadmium, the American Conference of Governmental Industrial Hygienists (ACGIH)recommends that urine cadmium concentrations should be below 5 µg/g creatinine (5). Themaximum urine concentration of cadmium detected in this investigation (1.76 µg/g creatinine) waswell below this health-based level.
A statistical summary of the concentrations of uranium that were detected in the urine samples inthis EI is presented in Table 3. None of the well water samples tested by the state laboratory had a uranium concentration in excess of the state drinking water.
|range||ND - 0.127||ND - 0.076|
ND = not detected
Most of the EI participants had urine uranium concentrations within the national reference range (95th percentile = 0.034 µg/g creatinine). However, the urine concentration of uranium was elevatedabove the 95th percentile comparison value in three participants. The Nuclear RegulatoryCommission has recommended that corrective actions be taken when urine uranium concentrationsin uranium mill workers exceed 15 µg/L . The maximum urine concentration of uraniumdetected in this investigation (0.127 µg/L) was well below this level.
A statistical summary of the concentrations of fluoride that were detected in the urine samples in thisEI is presented in Table 4. None of the well water samples tested by the state laboratory had afluoride concentration in excess of the state drinking water standard of 4 mg/L or the federal Secondary Maximum Concentration Level of 2 mg/L.
There is no national reference range for fluoride concentrations in urine. Fluoride concentrations inurine increase as the concentration of fluoride in a person's drinking water increases. In order toprevent dental caries, the U.S. Public Health Service (USPHS) recommends that public watersupplies be fluoridated at a concentration of 0.7 to 1.2 ppm, depending on the average maximumdaily temperature of the area (6). People who drink water that contains fluoride in this recommendedconcentration range typically have urine fluoride concentrations of about 1.0 mg/L (7, 8). The meanfluoride concentration in urine samples from the participants of this investigation was 0.96 mg/L.Therefore, the average fluoride concentration in the urine samples from the participants was similarto that in people who drink fluoridated water at the USPHS recommended levels.
Exposure to high doses of fluoride, particularly in young children, can cause dental fluorosis. As theconcentration of fluoride in drinking water increases above the recommended level, the risk andseverity of dental fluorosis increases. The U.S. Environmental Protection Agency set a secondarydrinking water maximum contaminant level (MCL) of 2.0 ppm fluoride in drinking water to protectagainst the cosmetic effect of dental fluorosis. All of the water concentrations of fluoride in theparticipants' wells were less than 1.0 ppm. People may also be exposed to fluoride from food,beverages, and fluoride-containing dental products. In the diet, high concentrations of fluoride arefound in tea, fish, and shellfish.
At higher concentrations of fluorine in drinking water (greater than 4.0 ppm), there is a risk ofskeletal fluorosis. In a study of workers with occupational exposure to fluoride, skeletal fluorosiswas observed in workers who had an average urine fluoride level of 9.0 mg/L (9). The AmericanConference of Governmental Industrial Hygienists (ACGIH) concluded that changes in bonestructure are unlikely to be seen if fluoride concentrations in 24-hour urine specimens are below 5mg/L (5). To protect workers who are exposed to fluoride in the workplace, the AmericanConference of Governmental Industrial Hygienists has recommended a Biological Exposure Index(BEI) of 3 mg/g creatinine prior to the start of the work shift and 10 mg/g creatinine at the end of the work shift (5).
The highest urine fluoride concentration detected in a participant of this investigation was 3.8 mg/Lor 2.3 mg/g creatinine. Therefore, fluoride exposures in the participants in this EI were not at levelsthat would be expected to cause skeletal fluorosis.
The summary statistics for the concentrations of boron detected in urine samples from theparticipants of this EI are shown in Table 5. There is no state or federal drinking water standard forboron. However, nine of the well water samples exceed the state health advisory level of 0.6 mg/L.
There is no national reference range for the concentration of boron in urine. In the publishedliterature, several studies reported urine concentrations of boron in populations with backgroundexposures to boron. The results of these studies are summarized in Table 6.
|Imbus ||United States||0.919||0.715||0.040-6.60|
|Usuda ||Japan||-||0.798 (GM*)||-|
|Abou-Shakra ||United Kingdom||-||0.753||0.155-2.888|
GM = geometric mean
The average boron concentration in the urine samples from the participants in this EI was 1.32 mg/L, and the median concentration was 0.98 mg/L. Therefore, as compared to the data inTable 6, the average/median urine concentration of boron in the urine samples from the participantswas similar to or slightly higher than those in populations with no unusual exposure to boron.
Imbus et al. reported that the 95th percentile urine concentration of boron in adult men in the UnitedStates was 2.46 mg/L . For the participants in this EI, urine concentrations of boron that exceedthis 95th percentile concentration were considered to be elevated. By this definition, seven of the EIparticipants had an elevated urine boron concentration. The maximum detected concentration ofboron in urine was 15.8 mg/L or 12.3 mg/g creatinine.
Even though several of the participants had elevated urine boron concentrations, the concentrationswere not at levels that have been associated with adverse health effects in humans or animals. In aborax mining area in Turkey, local residents were exposed to elevated concentrations of boron intheir well water, as well as through food . In this population, urine boron concentrations rangedfrom 0.04 to 50.70 mg/L, and the average concentration was 8.3 mg/L. In spite of these elevatedboron exposures, no adverse health effects were reported.
In borax workers in the United States, the end-of-shift average urine concentrations of boron inhealthy workers ranged from 3.16 to 10.72 mg/g creatinine . By comparison, in this EI, theaverage urine concentration of boron was 1.17 mg/g creatinine, and the maximum detectedconcentration was 12.3 mg/g creatinine. In experimental studies, adverse health effects were notobserved in animals that were exposed to boron at doses that greatly exceeded occupationalexposures. For example, no adverse health effects were observed in dogs that had been fed a dietcontaining boron that resulted in a urine boron concentration of 180 mg/L .
Therefore, although several participants in this investigation had urine boron concentrations thatwere elevated as compared to normal populations, the concentrations were not at levels that havebeen associated with adverse health effects.
At the time that this report was written, the NCEH laboratory had not yet analyzed the urine samplesfor speciated arsenic. When these results become available, ATSDR will prepare an addendum to this report that discusses the arsenic test results and their significance.
The participants in this investigation varied considerably in their use of well water. Some reportedusing well water for drinking, cooking, and non-potable purposes in their household. Others usedwell water only for non-potable purposes and used bottled water for drinking. Some of theparticipants reported using water treatment devices such as water softeners, particulate filters, carbonfilters, and reverse osmosis units. Furthermore, several of the participants reported that they stoppeddrinking the water when they received their water test results from the State, which was a day or twobefore this EI was conducted. In addition, whenever possible, the State collected water samplesprior to its passage through treatment devices. Therefore, the concentrations of chemicals in the rawwater could differ from what the participants were drinking from the tap.
Because of these multiple sources of variability, it is difficult to estimate exposures to drinking watercontaminants. Furthermore, this investigation was offered as a public health service, so we did notexclude any residents whose wells had been tested, regardless of their water consumption historyor use of water treatment devices. Several of the chemicals in this investigation (uranium, fluoride,and boron) are rapidly excreted from the body after ingestion, so the urine concentrations of thesechemicals are strongly influenced by recent consumption. Some of the participants in thisinvestigation had stopped drinking the water a day or two before they donated a urine sample. Theurine levels of contaminants that were measured in residents who were not drinking the water at the time of the investigation may not be representative of their past exposures.
In December 2003, individual test results and an explanation of their significance were provided to the participants of this investigation.
At the time of this investigation, lead, cadmium, uranium, fluoride, and boron were not detected inurine samples from the participants at concentrations that are associated with adverse health effects. Therefore, the measured exposures to these chemicals pose no apparent public health hazard. Thisconclusion may not apply to exposures that occurred in the past or to other residents of the area.
Blood lead concentrations are a better indicator of potential health effects than are urine leadconcentrations. Therefore, as a prudent public health policy, it is recommended that a blood leadmeasurement be conducted on the participants who had a urine lead concentration in excess of the 95th percentile comparison range.
- The Hillsborough County Health Department will offer blood lead testing to participantswho had a urine lead concentration in excess of the 95th percentile comparison range.
- The Florida Department of Health will prepare several health consultations for the CoronetIndustries site. These reports will assess all available data and information for the site and determine if environmental contamination poses a public health hazard.
Kenneth Orloff, PhD, DABT
Agency for Toxic Substances and Disease Registry
Michael L. Patterson, MD
Environmental Health Scientist
Agency for Toxic Substances and Disease Registry
Susan A. Bland
Florida Department of Health
Bureau of Environmental Community Health
- Florida Department of Health; Brief Petition Scoping Report: Coronet Industries; Plant City, Florida; June 18, 2003.
- National Center for Environmental Health; Second National Report on Human Exposure toEnvironmental Chemicals. http://www.cdc.gov/exposurereport/pdf/SecondNER.pdf
- BL Gulson et al. Blood lead-urine lead relationships in adults and children. EnvironmentalResearch, Section A 78 152-160 (1998).
- A Brockhaus et al. Exposure to lead and cadmium of children living in different areas ofNorth-west Germany: results of biological monitoring studies 1982-1986. Int Arch OccupEnviron Health 60 211-222 (1988).
- American Conference of Governmental Industrial Hygienists; Documentation of theBiological Exposure Indices; Cincinnati, Ohio.
- Centers for Disease Control, Recommendations for using fluoride to prevent and controldental caries in the United States. MMWR August 17, 2002/50(RR14); 1-452.
- Agency for Toxic Substances and Disease Registry, Draft Toxicological Profile for Fluorine,Hydrogen Fluoride, and Fluorides; September 2002.
- I. Zipkin et al. Urinary fluoride levels associated with use of fluoridated waters, PublicHealth Reports 71(8) 767-772 (1956).
- NL Kaltreider et al.; Health survey of aluminum workers with special reference to fluorideexposure. J Occup Med 14:531-541 (1972).
- C Minoia et al.; Trace element reference values from tissues from inhabitants of theEuropean community. I. A study of 46 elements in urine, blood, and serum of Italiansubjects. Sci Total Environ 95 89-105 (1990).
- HR Imbus et al. Boron, cadmium, chromium, and nickel in blood and urine. Archives ofEnvironmental Health 6 286-295 (1963).
- K Usuda et al. Study on urine boron reference values of Japanese men: Use of confidenceintervals as an indictor of exposure to boron compounds. Science Total Environment 22045-53 (1998).
- FR Abu-Shakra et al. Lithium and boron in biological tissues and fluids. Trace Elementsin Medicine 6 142-146 (1989).
- M Col and C Col. Environmental boron contamination in waters of Hisarick area in theKutahya province of Turkey. Food Chemical Toxicology 41 1417-1420 (2003).
- BD Culver et al. The relationship of blood- and urine-boron to boron exposure in boraxworkers and the usefulness of urine-boron as an exposure marker. Environmental HealthPerspectives 102(Suppl 7) 133-137 (1994).
- US Nuclear Regulatory Commission; NRC Regulatory Guide 8.22 - Bioassay at UraniumMills; 1988.