The Effect of Perchlorate, Thiocyanate, and Nitrate on Thyroid ...

0021-972X/05/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 90(2):700 –706 Copyright © 2005 by The Endocrine Society doi: 10.1210/jc.2004-1821

The Effect of Perchlorate, Thiocyanate, and Nitrate on Thyroid Function in Workers Exposed to Perchlorate Long-Term Lewis E. Braverman, XueMei He, Sam Pino, Mary Cross, Barbarajean Magnani, Steven H. Lamm, Michael B. Kruse, Arnold Engel, Kenny S. Crump, and John P. Gibbs Section of Endocrinology, Diabetes, and Nutrition (L.E.B., X.H., S.P.) and Departments of Radiology (M.C.) and Laboratory Medicine (B.M.), Boston Medical Center, Boston, Massachusetts 02118; Consultants in Epidemiology and Occupational Health (S.H.L., M.B.K., A.E.), Washington, DC 20007; Environ (K.S.C.), Ruston, Louisiana 71270; and Health Management Division (J.P.G.), Kerr-McGee Shared Services LLC, Oklahoma City, Oklahoma 73125 Perchlorate (ClO4ⴚ) and thiocyanate (SCNⴚ) are potent and nitrate (NO3ⴚ) a weak competitive inhibitor of the thyroid sodium-iodide symporter. To determine the effects of longterm, high ClO4ⴚ exposure on thyroid function, we conducted a study of 29 workers employed for at least 1.7 yr (50% over 5.9 yr) in an ammonium ClO4ⴚ production plant in Utah. Serum ClO4ⴚ, SCNⴚ, and NO3ⴚ; serum T4, free T4 index, total T3, thyroglobulin (Tg), and TSH; 14-h thyroid radioactive iodine uptake (RAIU); and urine iodine (I) and ClO4ⴚ were assessed after 3 d off (Pre) and during the last of three 12-h night shifts in the plant (During) and in 12 volunteers (C) not working in the plant. Serum and urine ClO4ⴚ were not detected in C; urine ClO4ⴚ was not detected in 12 of 29 and was 272 ␮g/liter in 17 Pre workers; serum ClO4ⴚ was not detected in 27 of 29 Pre; and serum and urine ClO4ⴚ were markedly elevated during ClO4ⴚ exposure to 868 ␮g/liter and 43 mg/g creatinine, respectively. Serum SCNⴚ and NO3ⴚ concentrations were similar in all groups. Thyroid RAIUs were markedly decreased in During

compared with Pre (13.5 vs. 21.5%; P < 0.01, paired t) and were associated with an increase in urine I excretion (230 vs. 148 ␮g I/g Cr; P ⴝ 0.02, paired t) but were similar to those in the C group (14.4%). Serum TSH and Tg concentrations were normal and similar in the three groups. Serum T4 (8.3 vs. 7.7 ␮g/dl), free T4 index (2.4 vs. 2.2), and total T3 (147 vs. 134 ng/dl) were slightly but significantly increased in the During vs. Pre workers (P < 0.01, paired t). Thyroid volumes and patterns by ultrasound were similar in the 29 workers and 12 community volunteers. In conclusion, high ClO4ⴚ absorption during three nights work exposure decreased the 14-h thyroid RAIU by 38% in ClO4ⴚ production workers compared with the RAIU after 3 d off. However, serum TSH and Tg concentrations and thyroid volume by ultrasound were not affected by ClO4ⴚ, suggesting that long-term, intermittent, high exposure to ClO4ⴚ does not induce hypothyroidism or goiter in adults. (J Clin Endocrinol Metab 90: 700 –706, 2005)

S

sphere. Preshift and postshift urine samples were collected, and ClO4⫺ and creatinine levels were measured. Shift absorbed doses were estimated assuming standard creatinine excretion rates and a serum ClO4⫺ half-life of 8 h. The estimated shift doses for the highest exposure group averaged 0.5 mg/kg䡠shift. There were no changes in thyroid hormone or TSH serum levels associated with long-term ClO4⫺ exposure at these levels. Lawrence et al. (3, 4) reported clinical studies in which healthy volunteers were administered 10 mg ClO4⫺/d (0.14 mg/kg䡠d) or 3 mg ClO4⫺/d (0.04 mg/kg䡠d). Thyroid radioactive iodine uptake (RAIU) was determined at baseline, after 2 wk of ClO4⫺, and 2 wk after ClO4⫺ had been discontinued. After 2 wk of ingesting ClO4⫺, RAIU had decreased from baseline measurements by 38 and 10% in the 10- and 3-mg dose groups, respectively. Two weeks after discontinuing ClO4⫺, RAIU was significantly elevated above baseline by 25 and 22%, respectively. There were no changes in any of the measured serum thyroid function tests (TSH, FTI, T3). Greer et al. (5) conducted a 2-wk ClO4⫺ ingestion study in healthy volunteers using doses of 0.5, 0.1, 0.02, and 0.007 mg/kg䡠d. After 2 wk of dosing, the 24-h RAIU was decreased by 67% at the highest dose and was essentially unchanged from baseline in the lowest-dose group. No associated in-

INCE THE DISCOVERY of the widespread presence of low concentrations of perchlorate (ClO4⫺) in drinking water sources in the southwestern United States in 1997, a number of studies have been conducted in an effort to assess the possible risks to human health. It has been theorized that chronic exposure to ClO4⫺ may cause a situation analogous to mild iodine deficiency, thus causing increased serum levels of TSH and thyroglobulin (Tg) and decreased serum levels of T3 and T4. Gibbs et al. (1) reported on a cohort of workers with longterm exposure to ammonium ClO4⫺. Airborne ClO4⫺ dust was measured, and work-shift doses were estimated based on standard breathing rates. Doses up to 0.45 mg/kg䡠shift were estimated. There were no changes in serum free T4 index (FTI) or TSH concentrations associated with either long-term exposure or single shift exposure. Lamm et al. (2) reported on the only other cohort of ammonium ClO4⫺manufacturing workers in the western hemiFirst Published Online November 30, 2004 Abbreviations: FTI, Free T4 index; RAIU, radioactive iodine uptake; Tg, thyroglobulin; TT3, total T3; T3U, T3 uptake. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

700

Braverman et al. • Thyroid Function and RAIU in Perchlorate Workers

creases in serum TSH or decreases in serum free T4 concentrations were found in this study. RAIU in the three higherexposure groups was approximately 5% above baseline 2 wk after ClO4⫺ was discontinued, but this increase did not reach statistical significance. Tonacchera et al. (6) determined the relative potencies of ClO4⫺, thiocyanate (SCN⫺), and nitrate (NO3⫺) in inhibiting RAIU in a cell-culture model and concluded that the effects were simply additive, with the inhibiting effect of any one of the three inhibitors being indistinguishable from a dilution or concentration of either of the remaining two. The relative potency of ClO4⫺ for inhibiting iodine uptake by the sodiumiodide symporter was 15 and 240 times greater than that of SCN⫺ and NO3⫺, respectively, on a molar concentration basis in serum. The current study was conducted in the same ammonium ClO4⫺ manufacturing facility that Lamm et al. (2) studied. The purpose of the current study was to determine whether long-term ClO4⫺ exposure leads to similar effects on the RAIU as the 2-wk clinical studies and whether the RAIU observed in vivo can be predicted from the cell-culture model of Tonacchera et al. (6). Subjects and Methods The protocol and informed consent were approved by the Georgetown University Institutional Review Board. The study was conducted at an ammonium ClO4⫺ production facility in Cedar City, UT. Workers in the facility work 12-h shifts on three consecutive days followed by three consecutive days off. They rotate shifts from days to nights monthly. Workers from four shifts were studied during separate weeks between January and April 2004 while they worked night shifts on Tuesday, Wednesday, and Thursday. The employees reported to the hospital on Monday evening and Tuesday morning preceding exposure (pre), the Thursday evening of the exposure period (intermediate) and the Friday morning after the last night of exposure (during). The workers performed essentially the same tasks on each of the three shifts. Workers were surveyed to determine which jobs they worked each night during the study and for the frequency that they worked each of the various jobs during the preceding year. A schematic of the study design is shown in Fig. 1. Seven to eight ClO4⫺ workers and three nonexposed community volunteer controls

J Clin Endocrinol Metab, February 2005, 90(2):700 –706

701

were studied each week. Frozen serum and urine samples were shipped to the Boston Medical Center under chain of custody on the Monday after the sampling.

Thyroid ultrasound Each worker and community volunteer had a thyroid ultrasound performed at the hospital radiology department, and the thyroid volume was determined. The ultrasounds were read by Dr. Stephen Phillips in Cedar City and by Dr. Ewa Kuligowska-Noble at the Boston Medical Center, both of whom were unaware of the subjects’ status.

RAIU determinations An Atomlab 950 thyroid uptake system was set up on site at the hospital, and hospital technician training was conducted by Mary Cross of Boston Medical Center. 123I, flown in on Monday and Thursday mornings, was administered on Monday evening and Thursday evenings. The thyroid ␥-count was conducted 14 h after a 75-␮Ci dose of 123I had been administered in an oral capsule.

Laboratory tests ClO4⫺, SCN⫺, and NO3⫺ measurements were carried out in one author’s laboratory (L.E.B.) using HPLC. ClO4⫺ was first extracted from sera with ethanol and then diluted with deionized water. SCN⫺ and NO3⫺ were extracted from sera with acetonitrile and then diluted with deionized water. The prepared samples were analyzed with the Dionex DX-600 IC system. The analytical column was the AS16 (4 ⫻ 25 mm), and the detection was conducted with a suppressed conductivity, ASRS Ultra, 4-mm external water mode. The mobile phase was a gradient elution with potassium hydroxide in deionized water, using an EG 40 eluent generator. Thyroid hormone tests (T3, T4, FTI, and TSH) were carried out at the end of the study in the same assay at the Boston Medical Center laboratories by chemiluminescence using the Bayer Advia Centaur automated system (Bayer Healthcare, Tarrytown, NY). Normal ranges are as follows: T4, 4.5–10.9 ␮g/dl; T3 uptake (T3U), 22.5–37.0%; FTI, 1.0 – 4.0; TSH, 0.35–5.5 ␮U/ml; total T3 (TT3), 60 –181 ng %. The FTI index is the product of the T4 concentration and the T3U divided by 100. Tg was measured using chemiluminescence on the Nichols Advantage (Nichols Institute Diagnostics, San Juan Capistrano, CA). Normal range for serum Tg is 4 – 40 ng/dl. Intraassay coefficients of variation are as follows: TSH ⬍ 2.5%, T4 ⬍ 3.2%, T3U ⬍ 3.2%, TT3 ⬍ 3.2%, and TSH ⬍ 2.5%. Urinary iodine was measured using the Sandell-Kolthoff reaction for iodine modified by Benotti et al. (7) in the laboratory of one of the authors (L.E.B.). Complete blood count and serum chemistry analyses were conducted by Labcorp. The schedule of tests performed is shown in Table 1. All tests in this schedule were conducted with the exception of the Thursday urine ClO4⫺ and creatinine measurements on the six workers tested during the first week.

ClO4⫺ dose estimates Shift ClO4⫺ doses were estimated from serum ClO4⫺ concentrations using assumptions of an 8-h serum half-life and a volume of distribution of 30%. These values were derived by fitting a first-order model to the Greer et al. (5) data. Average doses on the Tuesday and Wednesday night shifts (shifts 1 and 2) were estimated using the difference between the intermediate exposure serum concentration and the preexposure serum concentration. The Thursday night (shift 3) ClO4⫺ dose was estimated using the difference between the during-exposure serum concentration and the intermediate serum ClO4⫺ concentration. Additionally, the Thursday night (shift 3) ClO4⫺ dose was estimated using the method described in Lamm et al. (2) and the intermediate and during-exposure urine ClO4⫺/creatinine values. FIG. 1. Schematic of study design. ClO4⫺ workers work three consecutive 12-h shifts followed by three consecutive 12-h shifts off. All workers were studied during a week when they started their three work shifts at 1900 h on a Tuesday evening.

Predicted RAIU based on ClO4⫺, SCN⫺, and NO3⫺ Using the relative potencies and model developed by Tonacchera et al. (6), and the measured serum ClO4⫺, SCN⫺, and NO3⫺ concentrations,

702



TABLE 1. Schedule of serum and urine laboratory tests Monday evening

Tuesday morning (preexposure)

Serum

None

Urine

None

ClO4⫺ SCN⫺ NO3⫺ T3, T4, FTI, TSH, Tg Chem 20 and CBC ClO4⫺ Iodide Creatinine

Thursday evening (intermediate)

Friday morning (during exposure)

ClO4⫺

ClO4⫺ SCN⫺ NO3⫺ T3, T4, FTI, TSH, Tg

ClO4⫺ Creatinine

ClO4⫺ Iodide Creatinine

CBC, Complete blood count. the during-exposure predicted RAIU as a percentage of baseline was calculated.

Statistical methods Unpaired t tests were used to compare measurements in controls to those in exposed workers. Paired t tests were used to compare the results from exposed workers made during different work periods (e.g. during exposure compared with preexposure). In a few cases, the data were log-transformed before analysis to make them more normal. If the variances in the two groups differed significantly, the Satterthwaite approximation for unequal variances was used. All tests were conducted using SAS 8.0 (SAS Institute Inc., Cary, NC), and all reported P values are two-sided.

Results ⫺

Of 40 total ClO4 workers at the plant, 29 volunteered to participate in the study. An additional 12 community controls were recruited. The workers and controls did not differ in age (33.6 vs. 39.5 yr; P ⫽ 0.09), height (71.1 vs. 70.8 inches; P ⫽ 0.75), or weight (204.8 vs. 201.3 pounds; P ⫽ 0.82). All subjects were male and Caucasian. Clinical serum chemis-

tries (data not shown) indicated that all subjects had normal kidney and liver function. Eight workers and two community controls smoked cigarettes. The minimum duration of employment in ClO4⫺ manufacture among the workers was 1.7 yr, and the median was 5.9 yr. The employee turnover rate at this ammonium ClO4⫺ production facility for the year 2003 was 0.6%, approximately half of the U.S. national average for the same period. Table 2 contains results of the RAIU determinations, the ClO4⫺, NO3⫺, and SCN⫺ levels in serum, and the urine ClO4⫺ levels. Workers’ 14-h RAIUs measured during exposure were significantly lower than their preexposure baseline values (13.5% vs. 21.5% of administered dosage; P ⬍ 0.01, paired t) and were similar to those of the community controls (14.4%; P ⫽ 0.64, t test). The increased serum SCN⫺ concentrations present in the workers and community controls who smoked cigarettes (see below) did not affect the thyroid RAIU values. ClO4⫺ was detected in sera of workers only while working

TABLE 2. RAIU and serum concentrations of ClO4⫺, NO3⫺, and SCN⫺ P values n

RAIU, measured Workers during exposure Workers preexposure Controls Serum ClO4⫺ (␮g/liter)b Workers during exposure Workers, intermediate Workers preexposurec Controls Serum NO3⫺ (␮g/liter) Workers during exposure Workers preexposure Controls Serum SCN⫺ (␮g/liter) Workers during exposure Workers preexposure Controls Urine ClO4⫺ (mg/g creatinine) Workers during exposure Workers, intermediate Workers preexposure Preexposure nondetectable Preexposure detectable Controls a

29 29 12

Mean

13.5% 21.5% 14.4%

13.1% 18.4% 14.5%

SD

7.1% 8.2% 3.9%

Comparison with control

Comparison with during

0.64a ⬍0.01

⬍0.01

29 22 29 12

838.4 310.6 2.0 0.0

358.9 152.5 0.0 0.0

1268.4 407.6 7.6

⬍0.01 ⬍0.01 0.16

⬍0.01 ⬍0.01

29 29 12

7926.8 7638.6 7200.6

7898.5 6956.0 6229.9

1895.9 2440.9 2402.7

0.31 0.60

0.55

29 29 12

3487.8 3304.0 3487.8

2350.3 2147.5 2350.3

2440.9 2666.1 3699.4

0.46a 0.60a

0.44

29 23 29 12 17 12

43.0 23.8 0.16 0.0 0.27 0.0

Data were log-transformed before analysis. Detection limit is 0.5 ␮g/liter. c The only two detectable were 31.71 and 26.83 ␮g/liter. b

Median

19.2 8.4 0.11 0.0 0.18 0.0

53.6 32.0 0.21 0.21

⬍0.01 ⬍0.01 ⬍0.01

⬍0.01 ⬍0.01


their night shifts, with the exception of two samples collected in the preexposure period. The mean serum ClO4⫺ concentrations were greater at the end of the third work shift (during, 838 ␮g/liter) than at the beginning of the shift (intermediate, 310 ␮g/liter) (P ⬍ 0.01). Urine ClO4⫺ measurements were made on the community controls and the preshift workers on Tuesday morning. ClO4⫺ was undetectable in all of the community controls and in 12 of the preshift workers. The mean urine ClO4⫺ levels in the 17 preshift workers whose samples tested positive were 0.27 mg/g creatinine. Urine ClO4⫺ measurements for the workers were also made on Thursday evening and Friday morning (just before and immediately after the third work shift). Urine ClO4⫺ excretion was significantly higher after the last night of exposure than before it (43.0 vs. 23.8 mg/g creatinine; P ⬍ 0.01, paired t). Serum NO3⫺ levels were similar in all three groups, during-exposure workers, preexposure workers, and controls, with means in the range of 7200 – 8000 ␮g/liter (0.30 ⱕ P ⱕ 0.60 in pairwise tests). Serum SCN⫺ levels were also similar in all three groups with means in the range of 3300 – 3500 ␮g/liter (0.44 ⱕ P ⱕ 0.66 in pairwise tests). Serum SCN⫺ levels did differ between smokers and nonsmokers. Serum SCN⫺ levels were higher in the 10 smokers (7551 ⫾ 2866 ␮g/liter) than in the 31 nonsmokers (2095 ⫾ 900 ␮g/liter), and this difference was highly significant (P ⬍ 0.001, t test). Figure 2 shows the ClO4⫺ dose estimates based on serum concentration differences between preexposure and just before the third shift (intermediate) and the ClO4⫺ dose estimates based on urine concentration differences from just before the third shift to just after the third shift. These dose estimates are compared with the shift 3 ClO4⫺ dose estimates based on serum concentration differences. A ClO4⫺ shift dose could not be calculated for one worker whose serum was undetectable for ClO4⫺ both pre- and during-exposure. The estimated shift dose based on urine concentration differences for another individual working a low-exposure job could not be plotted on a log-log plot (– 0.014 mg/kg䡠shift based on urine concentration differences vs. ⫹0.003 mg/kg䡠shift based on serum concentration differences). The shift dose estimates

FIG. 2. Comparison of shift ClO4⫺ dose estimates for shift 3 based on urine concentration differences and of shift ClO4⫺ dose estimates for shifts 1 and 2 based on serum concentration differences, with shift ClO4⫺ dose estimates for shift 3 based on serum concentration differences.


703

span three orders of magnitude, and the close correlation between dose estimates based on serum and urine concentration differences during the same time period is apparent, as is that based on serum concentration differences across the first two shifts and the third shift. Most of the employees performed the same tasks on all three shifts. A cumulative distribution of estimated average shift doses over the preceding year based on the worker survey of job tasks performed during the previous year and the average ClO4⫺ dose for each job task during this study is shown in Fig. 3. Half of the workers experienced average ClO4⫺ doses in excess of 0.33 mg/kg䡠shift over the year preceding this study. Figure 4 shows the 14-h RAIU measured during shift 3 as a percentage of the measured preexposure 14-h baseline and estimated ClO4⫺ dosage based on the sera ClO4⫺ levels before and after the third shift. Group mean data from the Lawrence et al. (3, 4) and Greer et al. (5) clinical studies are shown on the same plot. There is apparent similarity in the dose-response relationship among the four studies. The thyroid hormone and urinary iodine measurements are shown in Table 3. There were no differences in serum TSH and Tg values between the pre- and during-exposure workers or between the workers and the community controls. The during-exposure workers have a slight but statistically higher T3, T4, FTI, and urinary iodine excretion (␮g iodine/g creatinine) than they do in their preexposure state. The preexposure workers were not different from the community controls in any of the measures of thyroid function, but they did have a significantly lower iodine excretion. There was no difference in thyroid volume or pattern by ultrasound between the workers and community controls. Serum chemistries (not shown) indicated normal renal and hepatic function in all workers and community volunteers. In Fig. 5, the measured RAIU as a percentage of baseline is compared with the predicted, based on serum ClO4⫺, SCN⫺, and NO3⫺ concentrations in vitro data by Tonacchera et al. (6). Both the intercept and the coefficient are highly significant (P ⬍ 0.005).

704



FIG. 3. Cumulative frequency distribution of mean shift ClO4⫺ dose estimates for 29 workers, based an employee survey of the frequency of the various jobs worked over the preceding year and the mean ClO4⫺ dose observed for each specified job during this study.

Discussion ⫺

The similarities between estimated ClO4 doses based on serum concentrations across the first two shifts and across the third shift demonstrate that the exposures are relatively consistent for given jobs over a short time period. The close agreement between estimated ClO4⫺ doses for shift 3 based on serum concentrations and those based on urine concentrations further validates the dose estimates in Lamm et al. (2). The general agreement between doses estimated in this study and in that previous study indicates that doses are relatively consistent over time. The ClO4⫺ exposure pattern in the workers can best be characterized as long term and intermittent. Their median absorbed doses (0.33 mg/kg䡠shift ⫻ 3 shifts/6 d ⫽ 0.167 mg/kg䡠d) are approximately equivalent to drinking 2 liters of water containing 5000 ␮g/liter ClO4⫺. Although much higher than doses potentially achieved from environmental exposure, these absorbed doses are still significantly lower than the 600-1000 mg/d doses previously used to treat hyperthyroidism. The mean urine ClO4⫺ excretion among the 17 preshift workers who had detectable ClO4⫺ was 0.27 mg/g creatinine or approximately 400 ␮g/24 h. This is

FIG. 4. ClO4⫺ dose-response relationship for percent RAIU relative to baseline in the current study compared with data from two separate 14-d clinical studies in healthy adult volunteers. Estimated ClO4⫺ dose shown is for shift 3 based on serum concentration differences.

equivalent to the urine excretion that would result from drinking 2 liters of water daily containing 200 ␮g/liter ClO4⫺, which is significantly higher than ClO4⫺ levels present in most drinking water sources of concern in the United States. The ClO4⫺ doses in this occupational cohort decreased the 14-h thyroid RAIU by an average of 38% among duringexposure employees relative to their preexposure baseline. This is the same decrease that Lawrence et al. (3) had shown in normal volunteers ingesting 10 mg ClO4⫺ daily for 2 wk. Serum TSH, serum Tg concentrations, and thyroid volume by ultrasound were not affected by ClO4⫺, suggesting that that these large occupational exposures did not induce adaptive changes seen in adults before the development of hypothyroidism or goiter. The very small increases in serum T4, FTI, and TT3 concentrations observed during ClO4⫺ exposure remain unexplained but may be caused by the suggestion that decreased thyroid iodine content enhances the response of the thyroid to TSH. Thus, Brabant et al. (8) gave 900 mg ClO4⫺ daily for 4 wk to volunteers and reported that thyroid iodine content decreased as assessed by thyroid fluorescence scintigraphy with no change in thyroid size. Serum Tg values increased 2-fold and serum TSH concentrations decreased with no change in total serum T3 or T4 values. The close agreement in dose-response between the present study and the clinical studies by Lawrence et al. (3, 4) and by Greer et al. (5) lend credence to the dose estimates and RAIU measurements in the current field study. There was little difference in RAIU changes between pre- and during-exposure in our studies measured at 14 h and those in the clinical studies measured at 4, 8, and 24 h. We have demonstrated that the workers in this study have a significant decrease in RAIU as a result of their occupation and that this has not adversely affected thyroid function. Findings in the current study are also consistent with the rebound increase in RAIU noted in the studies by Lawrence et al. (3, 4) which could have contributed to the pre- and during-exposure differences observed. The workers’ RAIU was 60% higher in their preexposed state than during ClO4⫺ exposure, and during exposure, their RAIU was similar to that of the community controls. These findings are also con-



705

TABLE 3. Thyroid hormones, thyroid volume, and urinary iodide concentrations P values n

T3 (ng/dl) Workers during exposure Workers preexposure Controls T4 (␮g/dl) Workers during exposure Workers preexposure Controls FTI Workers during exposure Workers preexposure Controls TSH (␮IU/ml) Workers during exposure Workers preexposure Controls Tg (ng/ml) Workers during exposure Workers preexposure Controls Thyroid volume (ml) Workers Controls Urine iodine:creatinine ratio (␮g I/g creatinine) Workers during exposure Workers preexposure Controls a

29 29 12

Mean

147.0 134.4 130.2

Median

147.0 138.0 129.5

SD

26.2 22.6 24.1

Comparison with control

Comparison with during

0.06 0.59

⬍0.01

29 29 12

8.31 7.72 7.45

8.60 8.00 7.30

1.50 1.56 1.61

0.11 0.62

⬍0.01

29 29 12

2.44 2.25 2.27

2.44 2.16 2.22

0.29 0.37 0.48

0.27 0.92

0.01

29 29 12

2.21 2.02 3.09

1.80 1.51 3.00

1.61 1.72 1.89

0.14 0.05a

0.09a

29 29 12

22.26 23.50 21.07

22.60 18.30 16.25

12.93 17.81 18.36

0.81 0.7a

0.60

29 12

10.8 11.2

9.1 10.9

4.8 5.17

0.79

29 29 12

230 148 296

174 133 247

163 88 183

0.25a 0.02a

0.02a

Data were log-transformed before analysis.

sistent with a study by El Ghawabi et al. (9) in which workers with long-term cyanide exposure were found to have significantly higher RAIU compared with controls on the first day of their work week. This suggests that up-regulation of sodium-iodide symporter expression possibly occurs in humans with chronic goitrogen exposure, although this has not been conclusively demonstrated. The urinary iodine excretion among employees during ClO4⫺ exposure was approximately 55% higher than in the preexposed state. We find it unlikely that this represents a short-term dietary change but rather suggests that the thyroid may be concentrating less of the dietary iodide during

FIG. 5. Comparison of measured in vivo percent RAIU relative to baseline in ClO4⫺ workers with predicted percent RAIU relative to baseline using serum ClO4⫺, SCN⫺, and NO3⫺ concentrations and the in vitro derived relationship reported in Tonacchera et al. (6). Both the intercept and the slope are highly significant (P ⬍ 0.005).

ClO4⫺ exposure. Whether higher urinary iodine excretion in the community control volunteers compared with the preexposure workers may have contributed to their lower thyroid RAIU remains conjectural. Intermittent, high exposure to ClO4⫺ for many years in workers employed in an ammonium ClO4⫺ production plant did not induce goiter or any evidence of hypothyroidism because serum TSH and Tg concentrations were unaffected even though 123I uptakes were decreased during the work shift. These workers had serum and urine ClO4⫺ concentrations far in excess of that which could occur from environmental contamination of water supplies.

706



Acknowledgments We are grateful to AmPac for allowing us to conduct the study at their facility and we thank Ann Ott, R.N., from Work Med at Valley View Medical Center in Cedar City, UT, who kept the study on track and who retired during the last week of testing. Received September 14, 2004. Accepted November 17, 2004. Address all correspondence to: Lewis E. Braverman, Boston University School of Medicine, 88 East Newton Street, Evans Building, Room 201, Boston, Massachusetts 02118-2347. E-mail: Lewis.Braverman@ bmc.org. Address reprint requests to: John P. Gibbs, M.D., P.O. Box 25861, Oklahoma City, Oklahoma 73125. E-mail: [email protected]. Funding for this study was provided by the Perchlorate Study Group.

3.

4. 5.

6.

7.

References 1. Gibbs JP, Ahmad R, Crump KS, Houck DP, Leveille TS, Findley JE, Francis M 1998 Evaluation of a population with occupational exposure to airborne ammonium perchlorate for possible acute or chronic effects on thyroid function. J Occup Environ Med 40:1072–1082 2. Lamm SH, Braverman LE, Li FX, Richman K, Pino S, Howearth G 1999

8.

9.

Thyroid health status of ammonium perchlorate workers: a cross-sectional occupational health study. J Occup Environ Med 41:248 –260 Lawrence JE, Lamm SH, Pino S, Richman K, Braverman LE 2000 The effect of short-term low-dose perchlorate on various aspects of thyroid function. Thyroid 10:659 – 663 Lawrence J, Lamm S, Braverman LE 2001 Low dose perchlorate (3 mg daily) and thyroid function. Thyroid 11:295 Greer MA, Goodman G, Pleus RC, Greer SE 2002 Health effects assessment for environmental perchlorate contamination: the dose response for inhibition of thyroidal radioiodine uptake in humans. Environ Health Perspect 110:927– 937 Tonacchera M, Pinchera A, Dimida A, Ferrarini E, Agretti P, Vitti P, Santini F, Crump K, Gibbs J 2004 Relative potencies and additivity of perchlorate, thiocyanate, nitrate and iodide on the inhibition of radioactive iodide uptake by the human sodium iodide symporter. Thyroid 14:1012–1019 Benotti J, Benotti N, Pino S, Gardyna H 1965 Determination of total iodine in urine, stools, diets, and tissue. Clin Chem 11:932–936 Brabant G, Bergmann P, Kirsch CM, Kohrle J, Hesch RD, von zur Muhlen A 1992 Early adaptation of thyrotropin and thyroglobulin secretion to experimentally decreased iodine supply in man. Metabolism 41:1093–1096 El Ghawabi SH, Gaafar MA, El-Saharti AA, Ahmed SH, Malash KK, Fares R 1975 Chronic cyanide exposure: a clinical, radioisotope, and laboratory study. Br J Ind Med 32:215–219

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.