APPENDIX E
Detailed Information on Endocrine Studies of Fluoride
The tables that follow contain detailed information on the endocrine studies discussed in Chapter 8, including study design, exposure information, and reported effects. Exposure conditions and duration and fluoride concentrations are provided as given in the published articles. Many of the tables include estimates of exposure in units of mg/kg/day to aid in comparing studies. When possible, these estimates were made from information (e.g., intake rate of drinking water, body weight) given in the articles. Where such information was not available in a published article, the assumptions used to make the estimates are listed in footnotes to the tables. Note that for most of the human studies, the exposure estimates (mg/kg/day) are for typical or average values for the groups and do not reflect the full range of likely exposures.
TABLE E-1 Effects of Fluoride on Thyroid Follicular Cell Function in Experimental Animals
Species and Strain |
Exposure Conditions |
Fluoride Concentration or Dosea |
Exposure Duration |
Effects |
Reference |
Rats (Hebrew University albino, males; infants at start, 30-32 g) See also Table E-16 |
Drinking water |
0.55, 1, or 10 mg/L (0.055, 0.1, and 1 mg/kg/day)b |
9 months |
No significant differences in basal metabolic ratio, thyroid weight, radioiodine uptake, total blood iodine, protein-bound iodine, or urinary excretion. TSH not measured. |
Gedalia et al. 1960 |
Rats (females, 180-230 g) |
Gastric tube 0.2 or 2.2 µg/day iodine in diet |
750 µg/day in 1 mL water (3.3-4.2 mg/kg/day) |
2 months |
No effect of fluoride on body weight, weight of thyroid, total composition of iodinated amino acids, or amount of iodide present in the thyroid. No effect of fluoride on iodine excretion in the higher-iodine group. Decreased protein-bound iodine, T3, and T4 (low-iodine group). Decreased biogenesis of T3 and T4 following administration of131I (low- and high-iodine groups). TSH not measured. |
Stolc and Podoba 1960 |
Rats (Wistar, males; initial weight 170-230 g; 13 per group) |
Drinking water Dietary iodine, 0.45 µg/g feed (0.45 ppm) |
0, 0.1, or 1 mg/day (0, 0.43-0.59, or 4.3- 5.9 mg/kg/day) |
60 days |
Decreased plasma T3 and T4, decreased free T4 index, increased T3-resin uptake (all changes statistically significant except for the decrease in T3 for the group receiving 0.1 mg/day)c TSH not measured. |
Bobek et al. 1976 |
Cows (Holstein; various states of lactation, 9-13 cows from each of 9 herds) See also Table E-3 |
Feed supplements |
1-22 mg/kg F in feed (estimated) (approximate doses, 0.03-0.7 mg/kg/day)d |
Chronic |
Urinary fluoride ≥ 2.9 mg/L (range 1.04-15.7mg/L, average 5.13 mg/L). Decreased T3, T4, cholesterol and increased eosinophils with increasing urinary fluoride (adjusted for stage of lactation); serum calcium correlated with T3 and T4. Fluorosis herds (S1, C4, V3, B2) had lower T4 than herds W, B, M, G (P < 0.05). Feeding of iodinated casein to herd B2 for 3 weeks resulted in 100% increase in milk production, increased hematopoiesis, reduced eosinophils, increased serum calcium, decreased serum phosphorus, and increase in serum T4 from 3.4 to 14.1 µg/dL. TSH not measured. Bone fluoride: mean, 2,400 ppm in ash (range, 850-6,935, 22 specimens from 8 herds). |
Hillman et al. 1979 |
Rats (Wistar) See also Table E-16 |
Drinking water and diet |
Water: 0, 1, 5, 10, 50, 100, or 200 mg/L Diet: 0.31 or 34.5 ppm (0, 0.1, 0.5, 1, 5, 10, or 20 mg/kg/day from water and 0.025 or 2.8 mg/kg/day from feed)e |
54-58 days |
Elevated T3 and T4 in rats on 1 mg/L in drinking water and low-fluoride diet. Low T3 and normal T4 in rats on 1, 5, or 10 mg/L in drinking water and high-fluoride diet. Decreased TSH and GH in animals receiving 100 or 200 mg/L in drinking water. Full details not available. |
Hara 1980 |
Species and Strain |
Exposure Conditions |
Fluoride Concentration or Dosea |
Exposure Duration |
Effects |
Reference |
Rats (Wistar, 3-month-old, 200-400 g) |
Drinking water Animals were kept 21 days on a diet containing 0.15% PTU to deplete their thyroid glands of iodine and thyro-globulin. For the next 2 days, a low- iodine diet (0.04 µg/g) was fed, but no more PTU. During the next 6 days the rats were given sufficient iodine (1.5 µg of iodide/mL of drinking water, labeled with 0.1 µCi of 125I). Then fluoride was given as indicated. |
60 or 200 mg/L (6-20 mg/kg/day)b |
6 days |
Serum fluoride at end of experiment (µg/mL): 0.165 (controls), 0.246 (60 mg/L), and 0.576 (200 mg/L). No significant differences from control values for relative thyroid weight, iodine content of thyroglobulin, thyroidal content of organic iodine, or amounts of monoiodotyrosine, diiodotyrosine, T3, and T4. TSH not measured. |
Sieben- hüner et al. 1984 |
Cows (Holstein, females; age 5-6 months at start, 30 animals total) |
NaF added to feed Iodine intake not stated, presumably adequate |
30 or 50 ppm in feed Approx. 0.8 or 1.4 mg/kg/day at 30 weeks of age; approx. 0.5 or 0.8 mg/kg/day at 100 weeks of age |
Data reported through age 100 weeks |
Serum fluoride at age 88 weeks (mg/L): 0.06 (controls), 0.20 (30 ppm in feed), and 0.28 (50 ppm in feed). Urinary fluoride at age 88 weeks (mg/L): 1 (controls), 13 (30 ppm in feed), and 20 (50 ppm in feed). Bone fluoride at age 17 months (ppm in tail vertebra, means of groups of 5 animals): 352 and 453 (controls), 2,306 and 2,712 (30 ppm in feed), and 3,539 and 3,946 (50 ppm in feed). No significant differences from control values for T4 concentration and T3 uptake at ages 40, 56, 72, and 88 weeks.f TSH not measured. |
Clay and Suttie 1987 |
Rats (Wistar, males and females; 120 ± 19 g at start, 212 animals total) See also Table E-2 |
Drinking water Low or normal iodine |
10 or 30 mg/L in drinking water (1 or 3 mg/kg/day)b |
7 months |
10 mg/L and normal iodine: no significant effect (some decrease in serum T4 and T3). 30 mg/L and normal iodine: statistically significant decreases in T4, T3, thyroid peroxidase, 131I uptake, [3H]-leucine uptake, and thyroid weight. 10 mg/L and low iodine: abnormalities in thyroid function beyond those attributable to low iodine; reduced thyroid peroxidase; low T4, without compensatory transformation of T4 to T3. TSH not measured. |
Guan et al. 1988 |
Species and Strain |
Exposure Conditions |
Fluoride Concentration or Dosea |
Exposure Duration |
Effects |
Reference |
Mice (Kunmin, males; 288 animals in 9 groups of 32 each; 13-15 g at start) |
Drinking water (NaF) Iodine: low (0 µg/L); normal (20 µg/L); excess 2500 µg/L) Low-iodine, low-fluoride chow fed to all groups. |
Low, 0 mg/L; normal,0.6 mg/L; excess, 30 mg/L (0, 0.06, and 3 mg/kg/day)b |
100 or 150 days |
For iodine-excess groups, thyroid weight relative to body weight decreased significantly with increasing fluoride intake. For iodine-deficient groups, goiter incidence at 100 days was 18%, 40%, and 66% for low-, normal-, and high-fluoride groups, respectively; at 150 days, goiter incidence was 81-100%. Fluoride-excess groups at 100 days had elevated T4 with all concentrations of iodine intake and elevated T3 for iodine-deficient animals. Fluoride excess significantly inhibited radioiodine uptake in iodine-deficient and iodine-normal groups. Incisor fluorosis occurred only in the fluoride excess groups; severity was greater in the iodine-deficient animals. Bone fluoride in fluoride-excess animals was greater in iodine-deficient (means, 2,560-2,880 ppm ash) or iodine-excess animals (means, 2,140-2,380 ppm ash) than in iodine-normal animals (means, 1,830-2,100 ppm ash). TSH not measured. |
Zhao et al. 1998 |
Cattle near aluminum smelter in India |
Contaminated pasture from smelter emissions No information on iodine intake |
Not available |
Notavailable |
Skeletal and enamel fluorosis (58% of animals within 3 km of plant were affected). Significantly decreased concentrations of T3. Significantly increased concentrations of alkaline phosphatase, inorganic phosphorus, and creatinine. Urinary fluoride averaged 26.5 mg/L close to smelter. Full details not available. |
Swarup et al. 1998 |
Cattle, buffaloes, sheep, and goats in 21 villages in India (286 calves, 1,675 adult cattle, 290 adult buffaloes, 780 goats, 564 sheep) |
Drinking water No information on iodine intake |
1.5-4 mg/L in drinking water |
Native livestock present in relevant area since birth |
Prevalence of enamel fluorosis up to 75% (adult buffalo), 70% (adult cattle), or 100% (calves), depending on location; prevalence of skeletal fluorosis up to 37.5% (buffalo) or 29% (cattle), depending on location; no evidence of enamel or skeletal fluorosis in goats or sheep. No clinical evidence of goiter in any fluorotic animals. Animals not showing clinical signs of fluorosis were not examined for goiter. No measurements of any thyroid hormone parameters or TSH. |
Choubisa 1999 |
Mice (Wistar, adult females; about 30 g at beginning; fluoride was administered during pregnancy and lactation)g |
Drinking water (Iodine intake 0.720 ± 0.12 µg/g in diet) |
500 mg/L in drinking water (50 mg/kg/day to the mothers)b |
From day 15 of pregnancy to day 14 of lactation |
Body weight of pups at 14 days old was reduced 35%; 75% decrease in plasma T4 in pups; 17% decrease in cerebral protein in pups; histological changes in cerebellum in pups. TSH not measured. |
Trabelsiet al. 2001 |
Species and Strain |
Exposure Conditions |
Fluoride Concentration or Dosea |
Exposure Duration |
Effects |
Reference |
Cows (3 years old with chronic fluorosis, 10 controls without fluorosis, from different regions of Turkey) |
Drinking water Iodine intakes not specifically stated |
5.7-15.2 mg/L in drinking water (approximate doses, 0.7-1.8 mg/kg/day)h |
Lifelong |
Mean values of T4, T3, and PBI in fluorotic animals were below the normal ranges and also significantly less than in controls. Low concentrations of bioavailable iodine in fluorosis region might be a factor. TSH not measured. |
Cinar and Selcuk 2005 |
aInformation in parentheses was calculated from information given in the papers or as otherwise noted. bBased on water consumption of about 10% of body weight. cATSDR (2003) stated that an intermediate-duration minimal risk level (MRL) derived from this study of thyroid effects in rats would have been lower (more protective) than the chronic-duration MRL derived from a human study of bone effects (0.05 mg/kg/day). dBased on feed consumption of 16 kg/day (dry weight) and body weight of 500 kg. eBased on water consumption of about 10% of body weight and feed consumption of about 8% of body weight. fText says “triiodiothyronine uptake” and table says “thyroxine uptake.” Data for different treatment groups were not given. gIn many mammalian species, maternal fluoride exposures are not well reflected by fluoride concentrations in milk; therefore, the impacts of fetal exposure and of reduced milk production by the mothers must also be considered. hBased on water consumption of 60 L/day and body weight of 500 kg. ABBREVIATIONS: GH, growth hormone; PBI, protein-bound iodine; TSH, thyroid-stimulating hormone. |
TABLE E-2 Summary of Effects of Fluoride Exposure for Rats with Different Amounts of Iodine Intake (Means ± SD)
Group |
Body Weight, g |
Urinary Fluoride, mg/L |
Urinary Iodine, µg/24 hours |
131I Uptake, % at 24 hours |
Serum T4, µg/dL |
Serum T3, ng/dL |
TPO, G.U./100 g of body weight |
[3H] Leucine Uptake, cpm/10 mg |
Thyroid Weight, mg/g |
293 ± 57 |
1.23 ± 0.22 |
1.110 ± 0.226 |
47.37 ± 5.66 |
3.64 ± 1.45 |
70.65 ± 30.29 |
2.04 ± 0.22 |
1,808 ± 358 |
9.97 ± 3.52 |
|
|
294 ± 85 |
6.65 ± 0.91c |
1.215 ± 0.357 |
44.74 ± 5.14 |
3.02 ± 1.48 |
61.96 ± 26.02 |
1.98 ± 0.51 |
1,728 ± 790 |
9.58 ± 2.40 |
|
254 ± 68c |
8.16 ± 0.89c |
1.150 ± 0.87 |
42.73 ± 4.31c |
1.44 ± 0.39c |
43.00 ± 11.31c |
1.73 ± 0.24c |
1,258 ± 293c |
7.90 ± 2.37c |
289 ± 72 |
1.23 ± 0.26 |
0.095 ± 0.029c |
0.76 ± 0.70c |
95.81 ± 25.18c |
2.57 ± 0.44c |
2,252 ± 683c |
19.91 ± 11.23c |
||
|
308 ± 63 |
6.23 ± 0.88c |
0.099 ± 0.017c |
0.65 ± 0.57c |
68.05 ± 21.96 |
1.75 ± 0.21c |
1,804 ± 459 |
20.13 ± 22.10c |
|
aNormal iodine: 310 ng/g in diet; 8.2 ng/mL in drinking water. bFluoride: 1.856 ppm in diet; 0.4 mg/L in drinking water. cP < 0.01, compared with group 1 (control). dLow iodine: 20-62.5 ng/g in diet; deionized drinking water. eAlso statistically significant at 2 hours and 6 hours (P < 0.01, compared with group 1). fFluoride: 1.743 ppm in diet; deionized water. ABBREVIATIONS: cpm, counts per minute; G.U., guaiacol unit; TPO, thyroid peroxidase. SOURCE: Guan et al. 1988. Reprinted with permission; copyright 1988, Chinese Medical Association. |
TABLE E-3 Summary of Selected Findings for Fluoride-Exposed Dairy Cows
Herda |
Number Observed |
Urinary Fluoride, mg/Lb |
Serum T4, µg/dLc |
Serum T3, ng/dLd |
Plasma Calcium, mg/dLe |
W |
12 |
2.92 ± 0.52 |
4.60 ± 0.34 |
175 ± 7.2 |
10.1 ± 0.15 |
B |
12 |
5.37 ± 0.43 |
4.83 ± 0.19 |
168 ± 5.8 |
9.5 ± 0.11 |
M |
12 |
6.39 ± 0.92 |
5.30 ± 0.38 |
177 ± 8.4 |
9.6 ± 0.11 |
G |
12 |
6.33 ± 0.74 |
4.82 ± 0.28 |
159 ± 7.7 |
9.4 ± 0.15 |
P |
12 |
3.47 ± 0.47 |
— |
— |
9.3 ± 0.12 |
S1 |
12 |
6.29 ± 1.08 |
3.59 ± 0.26 |
126 ± 8.4 |
9.1 ± 0.17 |
C4 |
9 |
—f |
2.21 ± 0.54 |
— |
9.5 ± 0.14 |
V3 |
10 |
— |
3.35 ± 0.47 |
— |
9.5 ± 0.13 |
B2 |
13 |
— |
3.39 ± 0.42 |
— |
8.9 ± 0.12 |
aHerd identification as reported by Hillman et al. (1979). Enamel fluorosis and elevated bone fluoride were confirmed in herds S1, C4, V3, and B2. Cows were uniformly distributed throughout lactation in all herds. bW < all others (P < 0.05). cC4 < all others; S1, V3, B2 < W, B, M, G (P < 0.05). dS1 < W, B, M, G (P < 0.05). eB2 < M, W; S1, P, G < W (P < 0.05). f—indicates not measured or not reported. SOURCE: Hillman et al. 1979. Reprinted with permission; copyright 1979, Journal of Dairy Science. |
TABLE E-4 Effects of Fluoride in Drinking Water on Thyroid Follicular Cell Function in Humans
Study Population(s) and Type |
Fluoride Concentrationa and Exposure Duration/Conditions |
Iodine Status and Other Information |
Effects |
Reference |
India, 3 villages, 2,008 persons, all ages Ecologic study; cross-sectional; entire population of each village included |
5.4, 6.1, and 10.7 mg/L (means for the villages) Lifelong |
Iodine in drinking water: 14.4-175.3 µg/L (inverse relationship to fluoride concentration). Iodine from salt: 86 µg/day. Calcium in diet: 480 mg/day. Diet considered deficient in proteins, fats, calcium, vitamins A and C. |
Transient goiters in persons aged 14-17; associated with increased fluoride in water and with decreased iodine in water. |
Siddiqui 1960 |
Israel, 2,685 girls, ages 7-18 Ecologic study; cross-sectional; may have included all eligible subjects, but not specifically stated |
<0.1-0.9 mg/L Lifelong |
Iodine in drinking water: <2-100 µg/L. |
Endemic goiter associated with low iodine content of water, but not with fluoride content of water. |
Gedalia and Brand 1963 |
U.S., adults ages 18-60; 106 from Crisfield, Maryland (42% female); 109 from New York City (29% female) Ecologic exposure measure; cross-sectional; no information on subject selection |
0.09 mg/L in New York City 3.48 mg/L in Crisfield, Maryland ≥10 years exposure |
General iodine status not given. Chrisfield: the 3 individuals with the highest PBI concentrations were all on iodine medication for non-thyroidal disease, and one of the individuals with the lowest PBI had had a partial thyroidectomy for a thyroid cyst.b New York City: the individualwith the highest PBI was taking 3 grains of thyroid daily.b |
No differences in PBI. No gross thyroid abnormalities or gross evidence for thyroid disease. Mild or moderate enamel fluorosis in 75% of individuals from Crisfield. |
Leone et al. 1964 |
Study Population(s) and Type |
Fluoride Concentrationa and Exposure Duration/Conditions |
Iodine Status and Other Information |
Effects |
Reference |
Nepal, 648 persons in 13 villages with similar iodine concentrations in water, all ages Ecologic study; cross-sectional; samples represented about one-third of the population in each village (children presenting for inoculations plus accompanying adults) |
< 0.1 to 0.36 mg/L Lifelong |
Iodine in drinking water: ≤1 µg/L. Diet low in iodine; iodized salt not available. Calcium in water, 3-148 mg/L. Magnesium in water, 0.5-77 mg/L. Water hardness, 10-670 ppm (as CaCO3). |
Goiter prevalence (5-69%) positively associated with fluoride concentration (ρ = 0.74, P < 0.01). Goiter prevalence of at least 20% associated with fluoride concentrations ≥ 0.19 mg/L. Goiter prevalence also associated positively with water hardness (ρ = 0.77, P < 0.01), calcium (ρ = 0.78, P < 0.01) and magnesium (ρ = 0.83, P < 0.01). Effect of fluoride was independent of that of hardness. |
Day and Powell-Jackson 1972 |
India, 9 patients with moderate to severe skeletal fluorosis (6 males, 3 females), mean age 29 years; 5 control individuals (3 males, 2 females), mean age 31 years Case-control study; individual estimates of current fluoride intake, measurements of fasting plasma fluoride and urinary fluoride; incomplete information on selection of subjects and controls |
7.8-8.0 or 24.5-25.0 mg/L Current exposure to 0.8 and 1.8 mg/L in water for the 2 persons who had moved Lifelong 2 persons had moved to nonendemic areas 2 or 5 years previously Symptomatic for 10-15 years |
Iodine status not given |
PBI values all normal (4.2-5.8 µg/100 mL). No evidence of goiter or thyroid dysfunction. |
Teotia et al. 1978 |
Germany, 13-15 years old, males and females, 17 in low-fluoride group and 26 in high-fluoride group Ecologic exposure measure; cross-sectional; no information on subject selection; 2 of the original 19 in low-fluoride group excluded upon discovery of hyperthyroidism |
0.1-0.2 and 3 mg/L Lifelong |
Iodine status not given |
No significant differences in T3 uptake, T4, free T4 index, T3, reverse T3, thyroglobulin, TSH, thyroglobulin antibodies, or microsomal thyroid antibodies. Unexplained decrease in thyroglobulin in girls (31.3 ± 12.9 ng/mL in the low-fluoride group and 13.8 ± 4.3 ng/mL in the high-fluoride group); this difference is also reflected in the means for boys and girls combined. |
Baum et al. 1981 |
Ukraine, 2 cities with different water fluoride concentrations Ecologic exposure measure; cross-sectional; no information on subject selection |
Values not given |
Iodine status not given |
Iodine deficiency and “adaptive amplification of the hypophyseal-thyroid system” (increased TSH?) in residents with high fluoride in drinking water; increased incidence of “functional disturbance” of the thyroid, but no structural changes. Full details not available. |
Sidora et al. 1983 |
Study Population(s) and Type |
Fluoride Concentrationa and Exposure Duration/Conditions |
Iodine Status and Other Information |
Effects |
Reference |
Ukraine, 47 healthy persons (ages 19-59), 43 persons with hyperthyroidism (ages 18-58), and 33 persons with hypothyroidism (ages 20-55) Ecologic exposure measure; cross-sectional; no information on subject selection other than by thyroid status See also Table 8-6 |
Region I: 0.5-1.4 mg/L (mean, 1.0) Region II: 1.6-3.5 mg/L (mean, 2.3) Lifelong (permanent residents) |
Iodine status not given |
Among normal individuals,significantly increased serum TSH and thyroidal 131I uptake and significantly decreased serum T3 in Region II, although values still within normal ranges. Differences between Regions I and II not seen among thyroidopathy patients. No information on the prevalence of thyroid disease in the two regions. |
Bachinskii et al. 1985 |
China, children ages 7-14, 250 in Area A and 256 in Area B Ecologic exposure measure; cross-sectional; no information on subject selection |
Area A, 0.88 mg/L (enamel fluorosis, 20.80%) Area B, 0.34 mg/L (enamel fluorosis, 16.00%) Lifelong |
Iodine in drinking water (µg/L): Area A, 5.21; Area B, 0.96 Goiter prevalence: Area A, 91%; Area B, 82% |
Area A had higher TSH, slightly higher 131I uptake, and lower mean IQ than Area B. Area A also had reduced T3 and elevated reverse T3, compared with Area B. Urine fluoride (mg/L): Area A, 2.56; Area B, 1.34-1.61. |
Lin et al. 1991 |
India, 22,276 individuals in a single district, all ages Ecologic study; cross-sectional; subjects included 1% of total population and 5% of school children of randomly selected villages |
≥1 mg/L Enamel fluorosis prevalence ranged from 6.0% to 59.0% (12.2% overall) Lifelong |
Iodine in drinking water ≥ 10 µg/L Goiter prevalence ranged from 9.5% to 37.5% (14.0% overall) |
Significant positive correlation between prevalence of goiter and enamel fluorosis (r = 0.4926, P < 0.001). No significant correlation between water iodine concentration and goiter prevalence (r = 0.1443, P > 0.05). In regions with water iodine concentrations > 20 µg/L, goiter prevalence was significantly higher in regions with fluoride > 2 mg/L (27.8%) than in regions with fluoride < 2 mg/L (17.1%). No evidence for functional changes in thyroid activity associated with the presence of goiter. |
Desai et al. 1993 |
China, no details available Ecologic study; probably cross-sectional; no information on subject selection |
High fluoride, values not given (Enamel fluorosis in children, 72.9%) Lifelong |
High iodine, values not given |
Urinary fluoride: 2.08 ± 1.03 mg/L Urinary iodine: 816.25 ± 1.80 µg/L. Reduced 131I uptake rate, elevated serum TSH, with respect to controls. Prevalence of thyroid enlargement was 3.8% in adults and 29.8% in children, and of enamel fluorosis, 35.5% and 72.9%, respectively. |
Yang et al. 1994 |
Study Population(s) and Type |
Fluoride Concentrationa and Exposure Duration/Conditions |
Iodine Status and Other Information |
Effects |
Reference |
India, 500 individuals from 52 villages in 2 districts; blood samples from randomly selected subset of control and fluorotic individuals Ecologic exposure measure; cross-sectional; no information on selection of original set of subjects |
1.0-6.53 mg/L (18 villages, <2 mg/L; 26 villages, 2-4 mg/L; 8 villages, >4 mg/L; 74% with slight to severe mottling of teeth) Control, 0.56-0.72 mg/L Lifelong |
Iodine status not given |
Serum fluoride (mg/L): 38%, <0.2; 47%, 0.2-0.4; 15%, >0.4. Significant increase in serum T4 (P < 0.001): 14.77 ±0.512 µg/dL versus 9.16 ± 0.63 µg/dLc (ranges, 7.2-20.0 versus 5.4-13.0). No significant differences in concentrations of serum T3 and TSH. |
Michael et al. 1996 |
South Africa, 671 children, ages, 6, 12, and 15, from six towns selected by fluoride concentration of drinking water Ecologic exposure measure; cross-sectional; study population included all children of the designated ages who spent their entire lives in the study towns See also Table E-5 |
Low: 0.3 and 0.5 mg/L Medium: 0.9 and 1.1 mg/L High: 1.7 and 2.6 mg/L Severe mottling of teeth in most children in the high-fluoride towns, not seen in the other towns Lifelong |
Iodine in water, 105 to > 201 µg/Ld Iodine in urine, 193 to > 201 µg/Ld (median values) Iodine status considered sufficient (possibly even high) |
Goiter prevalence ranged from 5.2% to 29.0% (15.3-29.0% for 5 of the 6 towns). The two towns with the highest fluoride had the highest goiter rates (27.7 and 29.0%). The town with 5.2% goiter prevalence had substantially less undernutrition than the other 5 towns. |
Jooste et al. 1999 |
India, 90 children, ages 7-18 with enamel fluorosis; 21 controls, ages 8-20 without enamel fluorosis Case-control study, subjects with and without enamel fluorosis, also selected by water fluoride concentration; cross-sectional; ecologic exposure measure (water fluoride concentration) but urine and serum fluoride also measured |
Children with dental fluorosis: 1.1-14.3 mg/L (mean, 4.37 mg/L) Children without fluorosis: Group I, 0.14-0.81 mg/L (mean, 0.23 mg/L); Group II, 0.14-0.73 mg/L (mean, 0.41 mg/L) Lifelong |
Iodine supplementation via iodized salt for more than a decade previously, considered satisfactory |
49 of 90 children with fluorosis had “well-defined hormonal derangements”; findings were borderline in the remaining 41 children. Five distinct categories of hormonal deviations: normal FT4 and FT3, elevated TSH (subclinical hypothyroidism, 23 of 90) normal FT4 and TSH, low FT3 (low T3 syndrome, 16 of 90); borderline low T3 in many of the other children normal FT4, elevated FT3 and TSH (7 of 90); T4 on low end of normal range, possible T3 toxicosis normal FT3, low FT4, elevated TSH (2 of 90) normal FT4, low FT3, elevated TSH (1 of 90) Categories 2-5 all associated with or can be caused by abnormal deiodinase activity. Only 4 control children had serum fluoride concentrations below the normal upper limit;approximately 50% of the control children also had “hormonal deviations”; children with “safe” water (< 1 mg/L fluoride) were taking in too much fluoride, presumably from nonwater sources. |
Susheela et al. 2005 |
Study Population(s) and Type |
Fluoride Concentrationa and Exposure Duration/Conditions |
Iodine Status and Other Information |
Effects |
Reference |
|
|
|
Urinary fluoride concentrations (normal upper limit, 0.1 mg/L): Children with fluorosis, 0.41-12.8 mg/L (mean, 3.96 mg/L) Controls, 0.09-4.2 mg/L Serum fluoride concentrations (normal upper limit, 0.02 mg/L): Children with fluorosis, 0.02-0.41 mg/L (mean, 0.14 mg/L) Controls, 0.02-0.29 mg/L |
|
aDue to the great range of ages included in the various studies, and because the reports do not include dose estimates (mg/kg/day), comparisons in this table are best made in terms of fluoride concentrations in drinking water. Approximations of representative doses have been made as follows: Day and Powell Jackson (1972) (iodine deficiency present): [F] = ≥ 0.2 mg/L; intake of 1 L/day for a 20-kg child; approximate dose ≥ 0.01 mg/kg/day. Bachinskii et al. (1985): [F] = 1.6-3.5 mg/L; intake of 2 L/day for a 70-kg adult; approximate dose, 0.05-0.1 mg/kg/day. Lin et al. (1991) (iodine deficiency present): [F] = 0.88 mg/L; intake of 1 L/day for a 30-kg child; approximate dose, 0.03 mg/kg/day. Michael et al. (1996): [F] = 1.0-6.5 mg/L; intake of 2 L/day for a 60-kg adult; approximate dose, 0.03-0.22 mg/kg/day. Jooste et al. (1999): [F] = 1.7 and 2.6 mg/L; intake of 1 L/day for a 20-kg child or 2 L/day for a 50-kg teenager; approximate doses, 0.09-1.3 mg/kg/day for the child and 0.07-0.1 mg/kg/day for the teenager. Susheela et al. (2005): [F] = 1.1-14.3 mg/L; intake of 2 L/day for a 50-kg teenager; approximate dose, 0.04-0.6 mg/kg/day. bMcLaren (1976) suggested that these individuals should not have been included in the samples or else that further research on the etiology should have been carried out. cThe units for serum T4 given by Michael et al. 1996 are ng/mL, but most likely µg/dL was meant. In units of µg/dL, these mean values are in the normal range for the controls and slightly above the normal range for the endemic fluorosis population. If the values are in ng/mL, then both means are below the normal range for serum T4. dIodine concentrations reported as 0.83 to > 1.58 µmol/L in water and 1.52 to > 1.58 µmol/L in urine. ABBREVIATIONS: FT3, free T3; FT4, free T4; PBI, protein-bound iodine. |
TABLE E-5 Summary of Selected Parameters for Six South African Towns
Town |
Sample Size |
Fluoride in Drinking Water, mg/L |
Goiter Prevalence, % |
Median Urinary Iodine, µg/La |
Iodine in Drinking Water, µg/Lb |
Iodine in Iodized Salt, ppm |
Williston |
85 |
0.3 |
15.3 |
> 201 |
105 |
28 |
Victoria West |
127 |
0.5 |
17.3 |
> 201 |
> 201 |
5 |
Frazerburg |
87 |
0.9 |
18.4 |
193 |
127 |
11 |
Carnarvon |
95 |
1.1 |
5.2 |
> 201 |
—c |
9 |
Brandvlei |
94 |
1.7 |
27.7 |
> 201 |
> 201 |
5 |
Kenhardt |
183 |
2.6 |
29.0 |
> 201 |
143 |
4 |
aReported as > 1.58, > 1.58, 1.52, > 1.58, > 1.58, and > 1.58 µmol/L, respectively. bReported as 0.83, > 1.58, 1.00, > 1.58, and 1.13 µmol/L, respectively. cNo water sample. SOURCE: Jooste et al. 1999. Reprinted with permission; copyright 1999, Macmillian Publishers Ltd. |
TABLE E-6 Summary of Findings in Healthy Persons and Persons with Thyroid Disease
Group |
Region |
No. |
Fluoride in Drinking Water, mg/L |
Fluoride in Urine, mg/L |
Fluoride in Urine, mg/day |
Fluoride in Serum, mg/L |
Fluoride in Erythrocytes, mg/L |
131I Uptake, 24 hours, % |
T4, µg/dLa |
T3, ng/dLb |
TSH, milliunits/L |
Hyperthyroid |
I |
21 |
1.2± 0.2 |
1.5± 0.2 |
2.1± 0.4 |
0.18± 0.01 |
0.46± 0.03 |
61± 7c |
19± 1.2c |
340± 46c |
0.8± 0.12c |
II |
22 |
2.2± 0.2d |
2.9± 0.5d |
3.9± 0.9d |
0.19± 0.01e |
0.51± 0.10 |
20± 1.8e |
460± 120e |
0.6± 0.08e |
||
Hypothyroid |
I |
14 |
1.1± 0.1 |
1.4± 0.2 |
1.6± 0.2 |
0.23± 0.02 |
0.55± 0.10 |
8.5± 2.7c |
2.0± 0.54c |
72± 26c |
51± 11c |
II |
19 |
2.5± 0.5d |
2.8± 0.4d |
3.7± 0.7d |
0.29± 0.02 |
0.61± 0.02 |
9.8± 1.3e |
2.3± 0.16e |
65± 6.5e |
58± 17e |
|
Controls |
I |
17 |
1.0± 0.1 |
1.5± 0.2 |
1.9± 0.3 |
0.21± 0.01 |
0.55± 0.10 |
24±3 |
7.5± 0.62 |
180± 20 |
2.4± 0.2 |
II |
30 |
2.3± 0.1c |
2.4± 0.2c |
2.7± 0.2c |
0.25± 0.01c |
0.61± 0.03 |
33± 4c |
7.3± 0.47 |
130± 13c |
4.3± 0.6c |
|
aReported as 250 ± 16, 261 ± 23, 26 ± 7, 29 ± 2, 97 ± 8, and 94 ± 6 nmol/L, respectively. bReported as 5.2 ± 0.7, 7.1 ± 1.8, 1.1 ± 0.4, 1.0 ± 0.1, 2.8 ± 0.3, and 2.0 ± 0.2 nmol/L, respectively. cP < 0.05 compared with controls residing in Region I. dP < 0.05 compared with patients with corresponding thyropathies residing in Region I. eP < 0.05 compared with controls residing in Region II. SOURCE: Adapted from Bachinskii et al. (1985). |
TABLE E-7 Effects of Clinical Fluoride Exposure on Thyroid Follicular Cell Function in Humans
Study Population(s) and Type |
Exposure Conditions and Duration |
Fluoride Concentration or Dose |
Effects |
Reference |
Switzerland, patients with hyperthyroidism, males and females, 15 total Clinical trial; nonblinded; comparison with before-treatment values; mechanistic rather than therapeutic study |
NaF, orally (3 times per day) or intravenously (once per day) Iodine status not given 20-245 days |
2-10 mg/day [0.029-0.14 mg/kg/day]a |
Clinical improvement in 6 of 15 patients (symptoms of hyperthyroidism relieved, both BMR and plasma PBI reduced to normal concentrations); BMR or PBI was often improved in the other 9 Greatest improvement in women between 40 and 60 years old with a moderate degree of thyrotoxicosis. |
Galletti and Joyet 1958 |
Germany, women with osteoporosis, 26 total completed 6 months of treatment (median age 62.1 years) Clinical therapeutic trial; nonblinded; comparison with before-treatment values; 38 patients originally enrolled, 3 excluded for disturbance of thyroid function |
NaF, orally (twice per day) Iodine status not given Only 10 patients took their medicine regularly (as indicated by measurements of plasma fluoride) 6 months |
36 mg/day or less Reduction to half dose necessary for 6 patients [0.3 or 0.6 mg/kg/day]b |
Tested for T3 uptake, T4, free T4 index, T3, and TSH; tested before start of trial and after 3 and 6 months. No changes observed in thyroid function or size. |
Eichner et al. 1981 |
Study Population(s) and Type |
Exposure Conditions and Duration |
Fluoride Concentration or Dose |
Effects |
Reference |
Denmark, osteoporosis patients, 140 females, 23 males, aged 16-84 years, mean 63.7 years Clinical therapeutic trial; non-blinded; 163 consecutive patients (1975-1983) presenting with osteoporosis and at least one atraumatic spinal fracture and who started treatment with fluoride, calcium and vitamin D; comparison with before-treatment values |
NaF, orally (3 times per day with meals) Iodine status not given Calcium phosphate and vitamin D were supplemented Mean duration 2.8 years (5 years for 43 patients) |
27 mg/day during first year Later adjusted to maintain serum fluoride between 0.095 and 0.19 mg/L (5 and 10 µmol/L) [0.45 mg/kg/day]b |
No changes in thyroid function (T4, T3, T3 uptake, TSH). Joint-related (51%) and gastrointestinal (25%) side effects at some point during treatment; 6% withdrew due to side effects; side effects rare when doses reduced to 14-18 mg/day. |
Hasling et al. 1987 |
aBased on 70-kg body weight. bBased on 60-kg body weight. ABBREVIATIONS: BMR, basal metabolic rate; PBI, protein-bound iodine |
TABLE E-8 Effects of Fluoride on Thyroid Parafollicular Cell Function in Experimental Animals
Species and Strain |
Exposure Conditions |
Concentration or Dosea |
Exposure Duration |
Effects |
Reference |
Rats (Sprague-Dawley, albino, 200 g at start, 16 total, both sexes) |
A: Drinking water (8 animals) B: Intraperitoneal (4 animals) C: Controls (4 animals) |
A: 40 mg/L [4 mg/kg/day]b B: 20 mg/kg/day |
A: 2 months B: 4 days (lived with controls for 2 months, ip injections on last 4 days) |
No morphological differences in parafollicular cells. No evidence for short-term release of calcitonin, but calcitonin not directly measured. |
Sundström 1971 |
Pigs (females, 20 with thyroidectomy at 10 weeks old, 20 intact; 8 months old at start of experiment; bred at 8 1/2 months old) |
Basal ration (Ca deficient); basal ration plus Ca and P; basal ration plus NaF; basal ration plus Ca, P, and NaF Iodinated casein (0.2 g/day) fed to thyroidectomized animals |
2 mg/kg/day (fluoride in ration adjusted periodically to maintain this dose) |
Approximately 6 months Experiment terminated when litters were 7 weeks old (maternal age approximately 14 months) |
Retarding effect on cortical bone remodeling; intact thyroid gland necessary for this effect (effect not seen in thyroidectomized animals with replacement of thyroid hormone but not calcitonin). Bone fluoride in intact animals (µg/g): basal, 285; basal plus Ca and P, 181; basal plus NaF, 3,495; basal plus Ca, P, and NaF, 3,249. Bone fluoride in thyroidectomized animals (ppm): basal, 280; basal plus Ca and P, 252; basal plus NaF, 3323; basal plus Ca, P, and NaF, 3197. |
Rantanen et al. 1972 |
aInformation in brackets was calculated from information given in the papers or as otherwise noted. bBased on water consumption of about 10% of body weight. |
TABLE E-9 Effects of Fluoride on Thyroid Parafollicular Cell Function in Humans
Study Population(s) and Type |
Exposure Conditions and Duration |
Concentration or Dosea |
Effects |
Reference |
India, 9 patients with moderate to severe skeletal fluorosis (6 males, 3 females), mean age 29 years; 5 controls (3 males, 2 females) mean age 31 years Case-control study; individual estimates of current fluoride intake, measurements of fasting plasma and urinary fluoride; incomplete information on selection of subjects and controls |
Drinking water, area with endemic skeletal fluorosis 2 persons had moved to nonendemic areas 5 or 2 years previously Exposed since birth Symptomatic for 10-15 years |
A) 8.7-9.2 mg/day for 3 persons (7.8-8.0 mg/L in water) [0.145-0.15 mg/kg/day]b B) 21.0-52.0 mg/day for 4 persons (24.5-25.0 mg/L in water) [0.35-0.87 mg/kg/day]b C) 2.5 and 3.8 mg/day for 2 persons (0.8 and 1.8 mg/Lin water) [0.04-0.06 mg/kg/day]b D) 1.2-2.2 mg/day for 5 controls (0.7-1.0 mg/L in water) [0.02-0.04 mg/kg/day]b |
Elevated calcitonin concentrations: A, 3 of 3; B, 4 of 4; C, 1 of 2 (8 of 8 individuals with intake ≥ 3.8 mg/day; plasma fluoride ≥ 0.11 mg/L (5.7µmol/L); urinary fluoride ≥ 2.2 mg/day). |
Teotia et al. 1978 |
Russia, description of subjects not available Occupational study; probably cross-sectional; full details not available |
Occupational exposure (fluorine production) Duration not available |
Not available |
Elevated concentrations of calcitonin in blood. |
Tokar’ et al. 1989 |
Review of epidemiological studies from 1963-1997 (45,725 children) See also Table E-12 |
Drinking water Comparison of groups with adequate (>800 mg/day) and inadequate (<300 mg/day) dietary calcium intake Exposed since birth |
1.5-25 mg/L |
Normal or elevated plasma calcitonin. |
Teotia et al. 1998 |
China, 50 male fluoride workers and 50 controls Occupational cohort study; cross-sectional; measurements of fluoride in serum and urine; full details not available |
Occupational exposure Duration not available |
Not available |
Elevatedconcentrations of serum calcitonin and parathyroid hormone. |
Huang et al. 2002 |
aDoses in brackets were calculated from information given in the papers; other information is as reported. bBased on 60-kg body weight. |
TABLE E-10 Summary of Selected Findings for Nine Patients with Endemic Skeletal Fluorosis and Five Controls
Case Numbera |
Age |
Sex |
Fluoride in Drinking Water, mg/L |
Fluoride Intake, mg/day |
Urinary Fluoride, mg/day |
Plasma Fluoride, mg/Lb |
Urinary Calcium, mg/day |
Plasma Calcium, mg/dL |
Calcitonin, µg/L |
IPTHc, µg/mL |
1control |
35 |
F |
1.0 |
1.2 |
0.8 |
0.023 |
120 |
9.5 |
< 0.08 |
< 0.35 |
3control |
22 |
M |
0.8 |
1.6 |
0.2 |
0.021 |
115 |
10.0 |
< 0.08 |
0.40 |
2control |
25 |
M |
0.8 |
1.8 |
0.6 |
0.030 |
95 |
10.2 |
< 0.08 |
0.50 |
4control |
32 |
M |
0.7 |
2.0 |
1.0 |
0.020 |
170 |
9.6 |
< 0.08 |
< 0.35 |
5control |
34 |
F |
1.0 |
2.2 |
1.2 |
0.038 |
130 |
9.8 |
< 0.08 |
0.35 |
2* |
25 |
M |
0.8 |
2.5 (38)d |
1.2 |
0.036 |
85 |
10.1 |
< 0.08 |
0.55 |
4* |
18 |
M |
1.8 |
3.8 (30)e |
2.2 |
0.12 |
80 |
9.7 |
0.14f |
0.40 |
8 |
36 |
M |
7.8 |
8.7 |
3.2 |
0.15 |
65 |
8.9 |
0.70h |
|
7 |
25 |
F |
8.0 |
9.2 |
4.2 |
0.15 |
60 |
8.3 |
0.10f |
0.50 |
6 |
22 |
M |
8.0 |
9.2 |
5.8 |
0.18 |
70 |
8.8 |
0.12f |
0.35 |
1 |
36 |
F |
24.5 |
21.0 |
10.0 |
0.11 |
75 |
9.8 |
0.18f |
0.40 |
3i |
34 |
F |
25.0 |
28.0 |
11.0 |
0.17 |
70 |
9.65 |
0.18f |
1.10h |
5 |
35 |
M |
25.0 |
48.0 |
15.0 |
0.14 |
65 |
9.8 |
0.10f |
0.80h |
9i |
58 |
M |
25.0 |
52.0 |
18.5 |
0.26 |
78 |
10.6 |
0.10f |
1.50h |
aCase number as reported by Teotia et al. (1978), arranged in order of increasing fluoride intake. Control subjects are indicated. Asterisks by the case numbers indicate patients no longer living in the high-fluoride area; case 2 had moved 5 years previously and case 4 had moved 2 years previously. bPlasma fluoride reported in µmol/L as follows: 1.2, 1.12, 1.6, 1.05, 2.0, 1.9, 6.1, 7.8, 8.0, 9.7, 5.7, 9.2, 7.5, and 13.6. cPlasma immunoreactive parathyroid hormone. dFluoride intake before moving had been 38 mg/day. eFluoride intake before moving had been 30 mg/day. fConsidered elevated above calcitonin concentrations found in normal controls. gListed as “<0.10” in Table 1 of Teotia et al. (1978) but assumed to be a misprint of “0.10” based on information in the text of that paper. hConsidered elevated above IPTH concentrations found in normal controls. iPatient had radiographic findings suggestive of secondary hyperparathyroidism. SOURCE: Adapted from Teotia et al. (1978). |
TABLE E-11 Effects of Fluoride on Parathyroid Function in Experimental Animals
Species and Strain |
Exposure Conditions |
Concentration or Dosea |
Exposure Duration |
Effects |
Reference |
Sheep (4 pairs of twin lambs) |
Drinking water No information on dietary calcium |
200 mg/L (NaF) [90 mg/L] [9 mg/kg/day]b |
1 week or 1 month |
After 1 week, only slight changes in parathyroid ultrastructure; after 1 month, hypertrophy and ultrastructural changes considered to be indicative of increased activity in most cells. Fivefold increase in blood PTH as early as 1 week, remained raised through 1 month. Severely reduced skeletal growth, no evidence of increased resorption, no definite pathology of kidney. |
Faccini and Care 1965 |
Rabbits (strain and sex not stated, 48-42 days old at start) |
Oral supplement No information on dietary calcium |
10 mg/kg/day |
14 weeks; some animals followed for another 24 weeks after withdrawal of fluoride |
No significant differences in serum calcium or magnesium; no significant differences in histological, morphometric, or ultrastructural features; no evidence for increased production of PTH or secondary hyperparathyroidism. PTH concentrations not measured. |
Rosenquist and Boquist 1973 |
Rats (Sprague-Dawley, weanling male, 45 g; either thyroid-parathyroidectomized or sham-operated; 17-21 animals per group) |
Drinking water 0.6% calcium in diet |
90 mg/L [9 mg/kg/day]b Controls, <1 mg/L |
15 days |
No effect of fluoride on serum calcium, serum phosphorus, or body weight in either group. No effect of fluoride on serum immunoreactive PTH in sham-operated group. Significantly increased periosteal bone formation, significantly decreased endosteal bone formation, increased endosteal bone resorption; effects on bone were thought not to be due to increased PTH activity. |
Liu and Baylink 1977 |
Rats (Sprague-Dawley, males, 290-300 g; 12 animals per group) |
Drinking water Dietary calcium not given |
150 mg/L [15 mg/kg/day]b |
10 weeks |
Ultrastructural evidence (from transmission electron microscopy) of increased parathyroid activity: higher percentage of active chief cells (90% versus 6%), increased numbers of secretory granules, accumulation of glycogen granules. Results considered indicative of a type of secondary hyperparathyroidism. |
Ream and Principato 1981a; 1981b; 1981c |
Rats (Wistar albino, males, 95-105 g; 5 animals per group) |
Intraperitoneal |
15.8 mg/kg (35mg/kg NaF) |
Single dose, killed 0-24 hours later |
Increased serum phosphorus; decreased urinary phosphorus; no change in serum calcium; increased urinary calcium; increased calcium, magnesium, and cAMP in renal cells (increase in cAMP was temporary); increased activity of Ca2+-ATPase in kidney. Effects were suppressed in thyroid-parathyroidectomized animals. PTH concentrations not measured. |
Suketa and Kanamoto 1983 |
Rats (Wistar, male, age 5 weeks, 80 g; 40 animals total) |
Drinking water and feed |
Drinking water: 50 mg/L in treated group, 0.5 mg/L in controls Feed: 5 mg/kg feed (0.26 mM/kg feed) [Approximate doses: treated group, 5.4 mg/kg/ day; controls, 0.45 mg/kg/day]c |
46 weeks Calcium-deficient diet for last 16 weeks (from age 35 weeks, approximately 500 g) for half of the animals |
Average serum immunoreactive PTH reduced in fluoride-treated animals (not significantly) at 35 weeks. At 51 weeks, normal increase in PTH in response to a dietary calcium deficiency did not occur in fluoride-treated animals (inhibition of normal parathyroid function). Small but significant increase in calculated cytoplasmic volume was observed in calcium-deficient animals given fluoride. Normal serum calcium concentrations in all groups. |
Rosenquist et al. 1983 |
Species and Strain |
Exposure Conditions |
Concentration or Dosea |
Exposure Duration |
Effects |
Reference |
Pigs (female, 8 months old, average weight 112 kg; 8 animals per group) |
Daily oral supplement High calcium and vitamin D in diet |
2 mg/kg/day (Fluoride in feed and water approximately 0.05 mg/kg/day) |
6 months (average weight, 166 kg) |
Plasma fluoride (mg/L): controls, 0.013; treated, 0.24; peak (40-100 minutes after dose), >1.9. Skeletal fluorosis without changes in plasma calcium, parathyroid activity, or vitamin D concentrations. No effect on PTH (measured after 4 months). |
Andersen et al. 1986 |
Sheep (females, 3 breeds, average age 6.0 ± 2.8 years, 55-60 kg; 2 groups of 7 animals) |
Oral with dry feed Normal dietary calcium without calcium supplementation |
0.45 or 2.3 mg/kg/ day (NaF 1 or 5 mg/kg/day)d |
45 days |
Significant decrease in serum calcium and phosphorus in both groups; significant increase in osteocalcin in second group. Variable increase in serum PTH in both groups, not statistically significant due to wide variation, but mean serum PTH in both groups at least twice as high at 45 days (4.9 ± 3.5 and 3.9 ± 0.9 milliunits/mL) as at beginning of experiment (1.9 ± 0.3 milliunits/mL in both groups). Effects on osteoblast birth rate and life span; increased bone formation and resorption, but formation greater than resorption (net increase in bone mass); possible secondary hyperparathyroidism. Serum fluoride (means, mg/L): initial (both groups), 0.10-0.11; final (45 days), first group, 0.24, second group, 0.82; peak > 0.5 at 3 hours after single dose of NaF at 3.5 mg/kg (fluoride, 1.6 mg/kg). Bone fluoride (means, ppm in ash): initial, 2,200-2,500; final, 2,700-3,200. |
Chavassieux et al. 1991 |
Rats (Sprague-Dawley, male, 40-50 g weanlings at start, 68-77 animals per group) |
Drinking water |
5, 15, or 50 mg/L (0.26-0.45, 0.69-1.31, and 2.08-3.46 mg/kg/day, decreasing with increasing body weight) |
3, 6, 12, or 18 months |
“No significant effect” on plasma calcium or alkaline phosphatase; specific data by treatment group not reported. PTH concentrations not measured. |
Dunipace et al. 1995 |
Rabbits (Dutch-Belted, female, 3 1/2 months old at start, 1.55 kg; 2 groups of 12 animals) See also Table E-16 |
Drinking water |
0 and 100 mg/L [7-10.5 mg/kg/day]e |
6 months |
Decreased serum calcium (3%, possibly in the protein-bound fraction). No statistically significant changes in PTH, vitamin D metabolites, or serum phosphorus; mean PTH elevated 3%. Increased bone-specific alkaline phosphatase and tartrate-resistant acid phosphatase, indicative of increased bone turnover. Increased bone mass, but decreased bone strength. Increased serum fluoride (0.73 mg/L versus 0.044 mg/L) and bone fluoride (6,650-7,890 ppm in ash versus 850-1,150 ppm in ash). High intake of calcium and vitamin D from rabbit chow, probable explanation for absenceof secondary hyperparathyroidism. |
Turner et al.1997 |
Rats (strain not available) |
Drinking water Dietary calcium adequate or low |
100 mg/L [10 mg/kg/day]b |
2 months |
Animals on low-calcium diet: osteomalacia, osteoporosis, accelerated bone turnover, increased serum alkaline phosphatase, increased osteocalcin, increased PTH. Animals on adequate calcium diet: slightly increased osteoblastic activity (elevated serum alkaline phosphatase activity and increased average width of trabecular bone after 1 year). |
Li and Ren 1997 |
Species and Strain |
Exposure Conditions |
Concentration or Dosea |
Exposure Duration |
Effects |
Reference |
Rats (Sprague-Dawley, male, 30 to40 g weanlings at start, 432 animals total) |
Drinking water Either calcium-deficient diet or diet deficient in protein, energy, or total nutrients |
5, 15, or 50 mg/L [0.5, 1.5, or 5 mg/kg/day]b |
16 or 48 weeks |
No significant effect on plasma calcium or alkaline phosphatase; specific data by fluoride treatment group not reported. PTH concentrations not measured. Calcium-deficient animals absorbed and retained more fluoride than controls and, in highest fluoride group, gained significantly less weight. Combination of general malnutrition and calcium deficiency was not examined. |
Dunipace et al. 1998 |
Monkeys (cynomolgus, females, 2.5-3.5 kg) |
Isoflurane anesthesia |
Not available |
2 hours |
Increased serum inorganic fluoride; decreased ionized calcium; increased PTH and osteocalcin in response to decreased calcium. Serum fluoride 0.070 mg/L versus 0.046 mg/L with ketamine/atropine anesthesia. |
Hotchkiss et al. 1998 |
Rats (Wistar, females, 4-5 months old, 130- 150 g) |
Drinking water |
500 mg/L |
60 days |
Hypocalcemia, attributed to suppressed gastrointestinal absorption of calcium. Decreased weight gain; inhibition of acetylcholinesterase and total cholinesterase in brain and serum; decreased spontaneous motor activity and endurance time. PTH not measured. |
Ekambaram and Paul 2001 |
Rats (Wistar, adult females, 150-170 g at start; fluoride administered during pregnancy and lactation)g |
NaF orally by feeding tube |
40 mg/kg/day NaF (18 mg/kg/day fluoride to the mothers) |
Day 6 of gestation through day 21 of lactation |
Hypocalcemia in mothers and offspring. PTH not measured. Significant changes in other serum cations (sodium, potassium) and phosphorus. Significant recovery on withdrawal of NaF. |
Verma and Guna Sherlin 2002b |
Rats (Sprague Dawley weanlings) |
Drinking water to dams and then to weanling pups Some groups with calcium deficient diet (dams and pups) |
50 mg/L (5 mg/kg/day)b |
Day 11 of gestation through 9 weeks old; continued until 15 weeks old with restored calcium, low fluoride, or both |
Decreased serum calcium, increased serum alkaline phosphatase, increased concentrations of vitamin D metabolites (both 25(OH)D3 and 1,25(OH)2D3). Decreased transcription of genes for vitamin D receptor and calbindin D 9 k; increased transcription of calcium-sensing receptor gene. Continued fluoride excess even with calcium supplementation continued to be detrimental. PTH not measured. |
Tiwari et al. 2004 |
aInformation in brackets was calculated from information given in the papers or as otherwise noted. bBased on water consumption of about 10% of body weight. cBased on water consumption of about 10% of body weight and feed consumption of about 8% of body weight; ATSDR (2003) gives a fluoride dose of 3.3 mg/kg (presumably per day) for these animals. dChoice of doses based on a therapeutic dose of NaF (1 mg/kg/day) and a toxic dose of fluoride (5 mg/kg/day) (Chavassieux et al. 1991). eBased on average daily water consumption of 163 mL, mean initial weight of 1.55 kg, and mean final weight of 2.33 kg for the fluoride-treated group. fThe dose was selected to produce toxic effects in a short time, without lethality (Ekambaram and Paul 2001). gIn many mammalian species, maternal fluoride exposures are not well reflected by fluoride concentrations in milk; therefore, the impacts of fetal exposure and of reduced milk production by the mothers must also be considered. |
TABLE E-12 Effects of Fluoride on Parathyroid Function in Humans (Clinical, Occupational, and Population Studies)
Study Population(s) and Type |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
Effects |
Reference |
Denmark, 14 normal subjects (5 fasting, 9 nonfasting, ages 22-38 years) Experimental study |
Oral dose of NaF |
27 mg of fluoride (60 mg NaF) [0.4 mg/kg]b Single dose Measurements made at 1, 2, 3, and 24 hours |
Decreased serum calcium and phosphorus; increased immunoreactive PTH. Measured serum fluoride peak 0.8-0.9 mg/L. Uncertainty as to peak fluoride and PTH, minimum Ca and phosphorus concentrations. No differences between fasting and nonfasting subjects except for a higher increase in serum fluoride concentration in fasting subjects. |
Larsen et al. 1978 |
France, 21 surgery patients (12 males and 9 females; ages 20-60 years) Experimental study; subjects had orthopedic (16), opthalmologic (3), or plastic (2) surgery; study excluded patients who were obese, had altered renal function or previously recognized diseases, or received blood transfusions or undescribed medications; initial values used as controls |
Enflurane anesthesia |
Not available 60-165 min. (mean, 95.5 ± 26 minutes) |
Variations in phosphorus clearance suggestive of a transitory hypersecretion of PTH; initial fall in serum calcium, return to preoperative concentration after 24 hours (variations in calcium balance were not highly significant). PTH not measured. Maximum serum inorganic fluoride: 0.12 mg/L (versus 0.039 mg/L in controls). |
Duchassaing et al. 1982 |
The Netherlands, 91 osteoporosis patients (61 females, 30 males; mean ages by type of treatment were 57.6-67.3 years) Clinical therapeutic trial; non-blinded; subjects had osteoporosis with one or more vertebral fractures before participation in the study, had normal concentrations for serum creatinine and liver enzymes, were treated as outpatients, were mobile and advised to exercise; pretreatment values used as controls |
Oral sodium fluoride (capsules, enteric coated tablets, or enteric coated, slow release tablets) Calcium supplementation of 1,000 mg/day |
Mean fluoride dosages by group between 18 and 36 mg/day (NaF, 40-80 mg/day) [fluoride, 0.57-1.1 mg/kg/day]b 2 years |
Patients divided into “responders” and “nonresponders” (NR) by (1) degree of increase in serum alkaline phosphatase concentration (20% NR); (2) changes in bone mineral content (26% NR); (3) occurrence of femoral neck fracture (6.6% NR). Patients with a fracture had lower serum alkaline phosphatase changes and higher increases in PTH. |
Duursma et al. 1987 |
England (7 healthy males; ages 24-43 years) Experimental study |
Oral NaF tablets Calcium intakes 400-800 mg/day |
27 mg/day (NaF, 60 mg/day) [fluoride, 0.39 mg/kg/day]b 3 weeks, followed up 6 weeks later |
No significant changes in plasma alkaline phosphatase, 25-hydroxy vitamin D, PTH, total and ionized calcium, phosphorus, or albumin. Significant increase in serum osteocalcin. PTH elevated slightly but not significantly (50 ± 17.6 pM/L after versus 43 ± 5.3 pM/L before); large standard deviation indicates variable response (not seen with other parameters). |
Dandona et al. 1988 |
Study Population(s) and Type |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
Effects |
Reference |
England, osteoporosis patients (34 females aged 49-74 years; 7 males aged 45-69 years; all with postmenopausal or idiopathic osteoporosis; all had normal renal function; 6 females were on hormone replacement therapy) Experimental study |
NaF orally in gelatin capsules Calcium supplementation was started at least 6 weeks (median, 20 weeks) prior to study |
27 mg/day (NaF, 60 mg/day) [fluoride, 0.39 mg/kg/day]b 8 days |
Decreased serum calcium (total and ionized); decreased serum phosphorus; increased concentrations of biologically active PTH (more than 5-fold); major changes occurred within 48 hours, some return toward normal after that. Patients divided into 2 groups by stability of serum calcium and phosphorus concentrations; the groups varied in their response to NaF with respect to mineral absorption and balance. |
Stamp et al. 1988 |
England, osteoporosis patients (22 controls; 2 males and 20 females, mean age 67 ± 8 years, range 51-83 years; 18 treated patients, 5 males and 13 females, mean age 61 ± 12 years, range 41-78 years; 10 patients were common to both groups [before and after treatment]; 8 females were on hormone replacement therapy) Experimental study; longitudinal for 10 patients |
NaF orally in gelatin capsules Calcium supplementation was started prior to study |
27 mg/day (NaF, 60mg/day) [fluoride, 0.39 mg/kg/day]b 15 ± 10 months |
Increased concentrations of biologically active PTH (bio-PTH) in treated group (log-transformed means, 10.6 versus 2.5 pg/mL; ranges, 1.6-126 versus 0.25-10.9 pg/mL). Significantly higher serum alkaline phosphatase (SAP) in treated group. Fluoride-treated patients with elevated concentrations of bio-PTH (> 18 pg/mL) had significantly lower concentrations of SAP than other treated patients, indistinguishable from controls; elevated bio-PTH also associated with relative hypophosphatemia and relative hypocalciuria; possibly excessive PTH accounts for “refractory” state of some patients—nonresponsiveness to fluoride therapy. |
Stamp et al. 1990 |
U.S., female osteoporosis patients (patients with previous history of hyperparathyroidism and several other conditions were excluded) Initial recruitment included 203 in-state patients from previous fluoride trials and 95 controls who had not taken fluoride; of these, 40 fluoride patients and 43 controls were scheduled for appointments; 15 fluoride patients were no longer taking fluoride or failed the appointments; 5 controls failed the appointments; final study included 25 fluoride patients and 38 controls (mean ages, 70.1 for fluoride group, 69.5 for controls) Cross-sectional study; fluoride-treated patients and non-treated controls recruited from database of osteoporosis patients of one investigator; fasting samples; analyses of drinking water, blood, and urine performed blindly; results reported as means of groups and as number outside the normal range for the parameter; urine and plasma fluoride were clearly different between groups; no significant difference in mean water fluoride concentrations See also Table E-17 |
Slow-release sodium monofluoro-phosphate plus calcium carbonate at 1,500 mg/day Most controls (n = 38) had calcium supplementation |
23 mg/day (mean dose) [fluoride, 0.33 mg/kg/day]b 1.4-12.6 years (mean, 4.2 years) |
No significant difference in mean calcium concentrations between groups; 2 of 25 individuals outside normal range (versus 0 of 38 controls). Significant difference (elevation) in mean alkaline phosphatase concentrations between groups; 8 of 25 individuals outside normal range (versus 0 of 38 controls); for those 8, a significant elevation in bone isoenzymes was found. For 24 of the 25 patients, calcium was significantly lower than baseline (pretreatment) values and alkaline phosphatase was significantly higher. PTH not measured. Urine fluoride (mg/L, mean and SD): fluoride group, 9,7 (4.1); controls, 0.8 (0.5); plasma fluoride (mg/L, mean and SD)c: fluoride group, 0.17 (0.068); controls, 0.019 (0.0076) |
Jackson et al. 1994 |
Study Population(s) and Type |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
Effects |
Reference |
China, healthy adults (approximately 120 per group, with either normal or inadequate nutritional intakes; mean ages of groups, 44.9-47.7 years) Cross-sectional cohort study; subjects grouped by location (water fluoride concentration) and nutritional status; populations generally similar (e.g., socially and economically); estimated fluoride intakes and measurements of urine and plasma fluoride and other parameters were made for individuals but results reported only for groups; probable overlap between low (< 0.3 mg/L) and middle (around 1 mg/L) fluoride exposure groups for each nutritional category; no mention of whether analyses were performed blindly See also Table E-17 |
Drinking water Normal nutrition defined as > 75 g of protein and >600 mg of Ca per day Inadequate nutrition defined as <60 g of protein and <400 mg of Ca per day |
0.23, 1.02, and 5.03 mg/L (normal nutrition) 0.11, 0.90, and 4.75 mg/L (inadequate nutrition) Estimated intakes: 1.70, 3.49, and 14.8 mg/day (normal nutrition); 1.20, 2.64, 15.32 mg/day (inadequate nutrition) At least 35 years of continuous residency in the study area |
Significant decrease in plasma calcium concentration associated with an increase in fluoride exposure in the populations with inadequate nutrition; not detected in subjects with normal nutrition. Elevated alkaline phosphatase activity with increased fluoride exposure in all populations, with higher values in subjects with inadequate nutrition. All valuesd within the normal range regardless of fluoride exposure and nutritional condition. PTH concentrations not measured. |
Li et al. 1995 |
U.S., osteoporosis patients (Group I, “good responders,” 13 postmenopausal females and 3 males; Group II, “poor responders,” 7 postmenopausal females and 3 males; Group III, untreated controls, 10 age-matched postmenopausal females) Cross-sectional study of fluoride-treated osteoporosis patients; non-fluoride-treated osteoporosis patients as controls |
Oral doses of NaF or sodium monofluorophosphate Calcium intake at least 1,500 mg/day |
30.6 ± 6.6 mg/day (range, 17.4-40.0 mg/day) [0.44 ± 0.9 mg/kg/ day; range, 0.25-0.57 mg/kg/day]b 32 ± 19 months (range, 13-89 months) |
Patients who showed a rapid increase in spinal bone density also showed a general state of calcium deficiency and secondary hyperparathyroidism. Serum PTH elevated in 4 “good responders” and 1 “poor responder” but no controls; all 5 with elevated PTH were calcium deficient; mean PTH concentrations were similar for all 3 groups. Degree of calcium deficiency in fluoride-treated patients was proportional to serum concentrations of PTH, alkaline phosphatase, procollagen peptide, and osteocalcin and to urine hydroxyproline concentrations. Fluoride therapy can cause calcium deficiency, even in patients with a high calcium intake; osteogenic response to fluoride can increase the skeletal requirement for calcium. |
Dure-Smith et al. 1996 |
U.S., 199 adult volunteers (mean ages of groups, 62.3, 58.6, 57.2 years) Ecological study; cross-sectional; subjects grouped by location (water fluoride concentration); subjects not randomly selected; nonfasting samples; urine and plasma fluoride concentrations significantly different for groups; study parameters reported by groups; no information on whether analyses were performed blindly See also Table E-17 |
Drinking water, natural fluoride Dietary calcium and calcium concentrations in drinking water were not discussed |
0.2, 1.0, 4.0 mg/L [0.003, 0.01, 0.06 mg/kg/day]b At least 30 years of continuous residency in their communities |
Some differences in mean plasma calcium and phosphorus concentrations among groups were statistically significant (lower calcium at 0.2 mg/L than 1.0 or 4.0; higher phosphorus at 4.0 mg/L than 0.2 or 1.0); no significant differences among mean alkaline phosphatase concentrations; all mean values were within normal ranges. PTH not measured. |
Jackson et al. 1997 |
Study Population(s) and Type |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
Effects |
Reference |
U.S., 75 osteoporosis patients (36 with placebo and 39 with fluoride) Placebo-controlled therapeutic study; subjects randomly assigned to treatment groups; no information on whether analyses were performed blindly |
Oral doses of slow-release NaF Both groups given calcium at 800 mg/day as calcium citrate |
23 mg/day (NaF, 50 mg/day) [approximate fluoride dose, 0.33 mg/kg/day]b 2 cycles of 12 months of treatment, 2 months off; analyses at 0, 6, 12, and 14 months for each cycle Calcium supplemented continuously throughout |
No significant changes in most parameters. Decrease in immunoreactive PTH from beginning values (due to increased calcium intake); fluoride-treated group slightly and consistently (but not significantly) higher than placebo group. Decrease in serum 1,25-dihydroxy vitamin D in placebo group but not in fluoride-treated group. |
Zerwekh et al. 1997b |
China, 50 male fluoride workers and 50 controls Occupational cohort study; cross-sectional; measurements of fluoride in serum and urine; full details not available |
Occupational exposure |
Not available |
Elevated concentrations of serum calcitonin and PTH. |
Huang et al. 2002 |
aInformation in brackets was calculated from information given in the papers or as otherwise noted. bBased on 70-kg body weight. cReported as 9.0 (3.6) µmol/L for the fluoride group and 1.0 (0.4) µmol/L for the controls. dNot stated whether this refers to mean values or all individual values. |
TABLE E-13 Effects of Fluoride on Parathyroid Function in Humans (Studies of Endemic Fluorosis Patients)
Study Population(s) |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
Effects |
Reference |
India, 25 cases of skeletal fluorosis (21 males, 4 females, aged 30-76, with radiologically proved skeletal fluorosis) 25 adult controls (19 males, 6 females, aged 25-75, not from endemic fluorosis area, and with no evidence of enamel or skeletal fluorosis or of bone or renal disease) Case-control study |
Drinking water (endemic fluorosis areas) |
Not given Probably lifelong |
No significant differences between cases and controls in serum calcium, serum inorganic phosphate, phosphate clearance, or 24-hour urinary calcium excretion (the latter either on a normal diet or on days 4-6 of a low-calcium diet); mean phosphate clearance was reduced, but not significantly. Significantly higher serum alkaline phosphatase values in individuals with fluorosis. No measurements of PTH. |
Singh et al. 1966 |
United States, 18-year-old boy, 57.4 kg, with renal insufficiency Case report See also Table 2-3 |
“High” intake of well water containing fluoride; current intake, 7.6 L/day (2 gallons per day) |
2.6 mg/L [0.34 mg/kg/day] Since early childhood |
Elevated serum immunoreactive PTH (more than 3 times normal value), slightly elevated serum calcium. Enamel fluorosis and roentgenographic bone changes consistent with “systemic fluorosis.”b |
Juncos and Donadio 1972 |
Study Population(s) |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
Effects |
Reference |
India, 20 patients with skeletal fluorosis (17 males, 3 females, age 42-68 years) Detailed studies on 5 of these patients (all males, age 42-60 years, duration of symptoms 5-11 years, no evidence of renal disease or intestinal malabsorption) Case reports; individual measurements of plasma and urine parameters and bone samples; comparison with values obtained from persons in nonfluorotic areas |
Drinking water (endemic fluorosis areas) Dietary calcium and vitamin D considered adequate |
> 25 mg/day [> 0.4 mg/kg/day]c Lifelong |
Clear evidence of secondary hyperparathyroidism in the 5 patients studied in detail; radiological findings consistent with hyperparathyroidism. Increased plasma alkaline phosphatase, increased phosphate clearance, decreased tubular reabsorption of phosphate, increased urinary fluoride, decreased urinary calcium. Normal plasma calcium and phosphate in 4 persons; elevated plasma calcium and decreased plasma phosphate in 1 person. Elevated serum immunoreactive PTH in all 5, especially in the person with elevated plasma calcium and decreased plasma phosphate (a parathyroid adenoma was later found in that individual, possibly attributable to long-standing hyperplasia as a result of excessive fluoride intake). Excess calcium and fluoride in bone in all 5 (11.8-13.2 versus 10.8 g of calcium per 100 g of dry fat-free bone ash; 265-585 versus 30 mg of fluoride per 100 g of dry fat-free bone ash). Urinary fluoride: 3.0-4.8 mg/L/day. |
Teotia and Teotia 1973 |
India, 9 patients with moderate to severe skeletal fluorosis (6 males, 3 females, mean age 29 years) 5 controls (3 males, 2 females; mean age 31 years) Case-control study; individual estimates of current fluoride intake, measurements of fasting plasma fluoride and urinary fluoride; incomplete information on selection of subjects and controls |
Drinking water, area with endemic skeletal fluorosis 2 persons had moved to non-endemic areas 5 or 2 years previously |
A) 8.7-9.2 mg/day for 3 persons (7.8-8.0 mg/L in water) [0.145-0.15 mg/kg/day]d B) 21.0-52.0 mg/day for 4 persons (24.5-25.0 mg/L in water) [0.35-0.87 mg/kg/day]d C) 2.5 and 3.8 mg/day for 2 persons (0.8 and 1.8 mg/L in water) [0.04-0.06 mg/kg/day]d D) 1.2-2.2 mg/day for 5 controls (0.7-1.0 mg/ L in water) [0.02-0.04 mg/kg/day]d Since birth Symptomatic for 10-15 years |
Increased PTH concentrations: A, 1 of 3; B, 3 of 4 [4 of 6 individuals with plasma fluoride ≥ 0.15 mg/L (7.8 µmol/L)]. Radiographs of 2 of the 4 persons were consistent with secondary hyperparathyroidism. |
Teotiaet al. 1978 |
India, 4 siblings (aged 8-18; 2 males, 2 females) and their mother (age 40), all with skeletal fluorosis Case reports; individual estimates of fluoride intake from water, measurements of serum fluoride and other parameters; age-matched Indian controls |
Drinking water source, 16.2 mg/L Calcium intakes considered normal (500-820 mg/day) |
16-49 mg/day from water, plus any contribution from food [0.5 mg/kg/day for the younger children; 0.5-1 mg/kg/day for the older children and mother]e Symptomatic for at least 2 years |
Normal total and ionized calcium concentrations; normal vitamin D concentrations in children; subnormal total and ionized calcium and subnormal vitamin D in the mother. Significantly elevated PTH, elevated osteocalcin, and elevated alkaline phosphatase in all 5. Findings consistent with secondary hyperparathyroidism. Skeletal changes, biochemical hyperparathyroidism, and elevated osteocalcin were similar in all 5, regardless of nutritional status (low in calories and protein for the mother, more nearly adequate for the children) and vitamin D status. Serum fluoride: 0.29-0.45 mg/L in the children (not measured in the mother). |
Srivastava et al. 1989 |
Study Population(s) |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
Effects |
Reference |
South Africa (260 children, 119 boys, 141 girls; ages 6-16, in an area with endemic skeletal fluorosis) 9 children (8 boys, 1 girl) studied individually; mean age, 13.7 ± 4.4 years; from the same area Prevalence (cross-sectional) study with ecologic measure of exposure; random selection of participants Case reports of 9 hospitalized individuals |
Drinking water |
8-12 mg/L [0.2-1.2 mg/kg/day]f Probably lifelong for most For the 9 children, at least 8 years |
Hypocalcemia present in 23% of the children; hypophosphatemia in 15%; elevated alkaline phosphatase in about 25%. Normal serum 25(OH)D concentrations in the 40 children in whom it was measured. Hypocalcemia in 6 of 9 studied individually; low concentrations of 25(OH)D in 2; elevated 1,25(OH)2D in 7. Bone fluoride elevated about 10-fold in the 7 children measured: 4,430-6,790 ppm in ash, mean 5,580 ppm in ash. Reduced phosphaturic response during a PTH-stimulation test (suggestive of pseudohypoparathyroidism Type II), directly related to presence of hypocalcemia, corrected by correcting the hypocalcemia. PTH concentrations not measured. Severe hyperosteoidosis associated with secondary hyperparathyroidism and a mineralization defect. Fluoride ingestion may increase calcium requirements and exacerbate the prevalence of hypocalcemia. |
Pettifor et al. 1989 |
Review of epidemiological studies from 1963-1997 (45,725 children) See also Table E-9 |
Drinking water Comparison of groups with adequate (> 800 mg/day) and inadequate (< 300 mg/day) dietary calcium intake |
1.5-25 mg/L Since birth |
High plasma fluoride, alkaline phosphatase, osteocalcin, PTH, and 1,25(OH2)D3; normal or elevated plasma calcitonin; normal plasma calcium, magnesium, phosphorus, and 25-(OH)D. Combination of fluoride exposure and calcium deficiency led to more severe effects of fluoride, metabolic bone diseases, and bone deformities. Toxic effects of fluoride occur at a lower concentration of fluoride intake (>2.5 mg/day) when there is a calcium deficiency; fluoride exaggerates the metabolic effects of calcium deficiency on bone. |
Teotia et al. 1998 |
India, children aged 6-12 in four regions (18-30 kg, 50 children per village) Cross-sectional cohort study; random selection of subjects; subjects grouped by location (water fluoride concentration); individual estimates of fluoride intake, measurements of serum and urinary fluoride, other end points; results reported by group See also Table E-14 |
Drinking water Calcium intake considered adequate (S.K. Gupta, Satellite Hospital, Banipark, Jaipur, personal communication, December 11, 2003) |
2.4, 4.6, 5.6, and 13.5 mg/L [0.25-0.41, 0.40-0.67, 0.48-0.80, and 1.1-1.8 mg/kg/day]g Lifelong |
Serum calcium concentrations within normal range for all groups; serum PTH concentrations elevated in two highest groups; serum PTH correlated with fluoride intake and with severity of clinical and skeletal fluorosis. |
Gupta et al. 2001 |
Study Population(s) |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
Effects |
Reference |
India, 1 adult female Case report |
Drinking water “8.4 times above the normal” |
Chronic |
Fluorosis, leading to secondary hyperparathyroidism manifesting as osteomalacia and a resorptive cavity in the head and neck of the femur; low serum calcium, elevated serum alkaline phosphatase; serum and urine fluoride “86 and 63 times above the normal.” |
Chadha and Kumar 2004 |
aInformation in brackets was calculated from information given in the papers or as otherwise noted. bJuncos and Donadio (1972) described two patients with renal insufficiency and systemic fluorosis; PTH was not reported for the second patient. cBased on consumption of 2 L of drinking water per day by a 60-kg adult. dBased on 60-kg body weight. eBased on 30- to 35-kg body weight for the younger children and 50- to 60-kg weight for the older children and mother. fBased on consumption of 1-2 L of drinking water per day by a 20-to 40-kg child. gBased on mean intakes (mg/day) for 18- to 30-kg children. ABBREVIATIONS: 25(OH)D, 25-hydroxy vitamin D; 1,25(OH)2D, 1,25-dihydroxy vitamin D. |
TABLE E-14 Summary of Selected Findings for Children in Four Villagesa
Village |
Fluoride in Drinking Water, mg/L |
Fluoride Intake, mg/dayb |
Serum Fluoride, mg/L |
Urinary Fluoride, mg/L |
Serum Calcium, mg/dL |
IPTHc, pM/L |
Enamel Fluorosis Scored |
Clinical Fluorosise |
Skeletal Fluorosisf |
Ramsagar ki Dhani |
2.4 |
7.35 (1.72) |
0.79 (0.21) |
9.45 (4.11) |
9.23 (1.89) |
31.64 (2.82) |
2.71 (1.09) |
0.95 (0.22) |
0.68 (0.67) |
Rampura |
4.6 |
11.97 (1.8) |
1.10 (0.58) |
15.9 (9.98) |
10.75 (1.66) |
40.98 (26.9) |
1.73 (1.09) |
1.00 (0.00) |
0.50 (0.61) |
Shivdaspura |
5.6 |
14.45 (3.19) |
1.10 (0.17) |
17.78 (7.77) |
9.68 (0.99) |
75.07 (31.75) |
2.44 (1.32) |
1.00 (0.00) |
0.79 (0.91) |
Raipuria |
13.6 |
32.56 (9.33) |
1.07 (0.17) |
14.56 (7.88) |
10.39 (1.44) |
125.10 (131.14) |
3.43 (1.70) |
1.51 (0.51) |
0.95 (1.12) |
aMean (standard deviation) of 50 children per village, ages 6-12, body weight 18-30 kg. bTotal from food and water. cPTH, midmolecule fragment; normal range, 48.1 ± 11.9 pM/L. dGrading of enamel fluorosis: 0, normal; 0.5 questionable fluorosis; 1, very mild fluorosis; 2, mild fluorosis; 3, moderate fluorosis; 4 severe fluorosis (defined in more detail by Gupta et al. 2001). eClinical (nonskeletal) fluorosis grading: 1, mild; 2, moderate; 3, severe (defined in more detail by Gupta et al. 2001). fSkeletal (radiological) fluorosis grading: 1, mild; 2, moderate; 3, severe (defined in more detail by Gupta et al. 2001). SOURCE: Gupta et al. 2001. Reprinted with permission; copyright 2001, Indian Pediatrics. |
TABLE E-15 Effects of Fluoride on Pineal Function in Animal and Human Studies
Species |
Exposure Conditions |
Concentration or Dosea |
Exposure Duration |
Mongolian gerbil (Meriones unguiculatus; males and females, from birth) |
Fluoride in feed (primarily); oral administration of fluoride through 24 days for high-fluoride group |
Low-fluoride group, 7 mg/kgfeed after age 24 days [0.7 mg/kg/day]b High-fluoride group, 2.3 mg/kg/ day orally, 5 days/ week through age 24 days; 37 mg/ kgfeed thereafter [3.7 mg/kg/day]b,c |
Birth through 28 weeks 24-hour urinary 6-sulfatoxymelatonin measured at 7, 9, 11.5, 16, 28 weeks |
Humans (female; 233 in Newburgh, NY; 172 in Kingston, NY) Ecologic study; most of the eligible children in both cities; nonblinded |
Fluoride in drinking water |
Newburgh, 1.2 mg/L [0.01-0.2 mg/kg/day]d Kingston, “essentially fluoride-free” [0.001-0.02 mg/kg/day]e |
Up to 10 years (ages 7-18 at time of study; ages at beginning of exposure varied from prenatal to 9 years) |
Humans (female; 337 in Kunszentmárton and 467 in Kiskunmajsa, ages 10-19.5 at time of study) Ecologic study; probably included most of the eligible children in both cities; nonblinded |
Fluoride in drinking water (probably natural fluoride) |
Kunszentmárton, 1.09 mg/L Kiskunmajsa, 0.17 mg/L [0.01-0.2 mg/kg/ day versus 0.001-0.02 mg/kg/day]f |
Lifelong |
aInformation in brackets was calculated from information given in the papers or as otherwise noted. bBased on estimated feed consumption of about 10% of body weight per day. cHigh-fluoride group was given 50 mg/L in drinking water during 24-hour metabolism studies when usual feed was not given. dEstimated fluoride intakes based on ranges of weight and water consumption for children aged 0-18 and fluoride concentration of 1.2 mg/L in drinking water; higher fluoride intakes are associated with the smallest children or the highest water intakes. Some individual intakes could have been lower or higher than the range shown. |
TABLE E-16 Effects of Fluoride on Other Endocrine Organs in Experimental Animals
Species and Strain |
Exposure Conditions |
Concentration or Dosea |
Exposure Duration |
Rabbits (young adult) |
Intravenous |
3 mg/kg/day |
2 months |
Rats (Long-Evans; 2 groups, each with 10 experimental and 5 control; age 49 or 52 days at start, 160-180 g) |
Intraperitoneal (controls injected with NaCl) |
Acute, 406.47 mg, NaF total [average dose, 68 mg/kg/day]b Chronic, 1131.65 mg of NaF total [average dose, 18 mg/kg/day]b |
Acute, 15 days Chronic, 100 days |
Rats (Hebrew University albino, males; infants at start, 30-32 g) See also Table E-1 |
Drinking water |
0.55, 1, or 10 mg/L [0.055, 0.1, and 1 mg/kg/day]c |
9 months |
Rats (Sprague-Dawley, males, 325-350 g) |
Intravenous |
6 mg/kg/hour |
3 hours |
Rats (Wistar) See also Table E-1 |
Drinking water and diet |
Water: 0, 1, 5, 10, 50, 100, or 200 mg/L Diet: 0.31 or 34.5 ppm [0, 0.1, 0.5, 1, 5, 10, or 20 mg/ kg/day from water and 0.025 or 2.8 mg/kg/day from feed]d |
54-58 days |
Rats (Wistar albino, males, 95-105 g) |
Intraperitoneal (controls injected with NaCl) |
15.8 mg/kg (35 mg/kg of NaF) |
Single dose |
Rats (inbred strain IIM, females, 180-220 g) |
Oral administration of NaF by gastric tube |
7.6 mg/kg |
Single dose, after fasting for 24 hours |
Effects |
Reference |
Adrenal weights averaged 20% greater than in controls. Body weight increase was 17% lower than in controls. |
Stormont et al. 1931 |
Acute: 7 of 10 survived, 6 were analyzed (1 “exhibited such bizarre overall changes” that it was omitted from the study). Chronic: 5 of 10 survived. Increased adrenal weight (about 30%) in both groups; enlarged adrenal cortex; normal cortical and medullary cytology. Increased width of connective tissue and increased mitotic activity in pancreases of most animals. |
Ogilvie 1953 |
No histological changes or weight differences in adrenals or pancreases; increase in pituitary weight (not significant for 1 mg/L, significant for 10 mg/L). |
Gedalia et al. 1960 |
Depression of glucose utilization, measured in terms of the output of 14CO2; serum glucose was not measured but presumably was elevated in accordance with decreased utilization. |
Dost et al. 1977 |
Decrease in pituitary weight in animals receiving 200 mg/L in drinking water. Decreased TSH and growth hormone in animals receiving 100 or 200 mg/L in drinking water. Full details not available. |
Hara 1980 |
Elevated serum glucose and enhanced glucose-6-phosphate dehydrogenase (G6PD) activities in liver and kidney; attributed to stimulation of adrenal function, both medullary and cortical; changes in glucose concentrations and G6PD activities suppressed by adrenalectomy but not by thyroid-parathyroidectomy. |
Suketa et al. 1985 |
Immediate fall in insulin concentrations (to 50% of basal concentration after 15 minutes) and consequent increase in glycemia (peak at about 1 1/2 hours), returned to normal in 4-5 hours. Decreased insulin response to glucose challenge when fluoride administered 15 minutes before glucose challenge (versus together with or immediately after). Appeared to be direct effect on insulin secretion, not on insulin receptors; hypoglycemic response to exogenous insulin was not impaired by pretreatment with fluoride. Plasma fluoride: 0.1-0.3 mg/L (5-15 µmol/L). |
Rigalli et al. 1990 |
Species and Strain |
Exposure Conditions |
Concentration or Dosea |
Exposure Duration |
Rats (female, IIM line, age 21 days at start) |
Drinking water (NaF) |
95 mg/L (5 mmol/L) [10 mg/kg/day]c |
100 days |
Rats (Sprague-Dawley, male, 40-50 g weanlings at start, 68-77 animals per group) |
Drinking water |
5, 15, or 50 mg/L [0.26-0.45, 0.69-1.31, and 2.08-3.46 mg/kg/day] (changing with increasing body weight) |
3, 6, 12, or 18 months |
Rats (female, IIM line, age 21 days at start) |
Drinking water (NaF) |
95 mg/L (5 mmol/L) [10 mg/kg/day]c |
3 months |
Rats (Zucker, males, normal and fatty diabetic, age-matched, 8 weeks old at start of study, initial weights 282 g for controls and 351 g for diabetics) |
Drinking water (NaF) (minimal contribution from feed) |
0, 5, 15, or 50 mg/L in drinking water (<1.2 ppm in feed) [Control: 0.05, 0.31, 0.85, and 2.8 mg/kg/day Diabetic: 0.09, 2.0, 6.0, and 15.5 mg/kg/day]e Reported doses for control rats (mg/kg/day): 0.33 for 5 mg/L and 3.04 or 50 mg/L; for diabetic rats, 1.99 for 5 mg/L and 16.26 for 50 mg/L |
3 or 6 months |
Effects |
Reference |
Subtle disturbance of glucose tolerance as shown by glucose tolerance tests, associated with period of elevated fluoride concentrations in plasma and soft tissue (deterioration of glucose tolerance for about 50 days and then normalization by 100 days, when maximum bone mass was achieved and plasma fluoride returned to normal concentrations). Bone mass higher 6-12% greater in fluoride-treated animals (depending on portion of skeleton considered). Bone fluoride (ppm in ash): controls, 1,160-1,410; treated, 6,880-8,550 (depending on portion of skeleton considered). |
Rigalli et al. 1992 |
“No significant effect” on fasting plasma glucose concentrations; specific data by treatment group not reported. |
Dunipace et al. 1995 |
Abnormal glucose tolerance tests when plasma diffusible fluoride exceeds 0.1 mg/L (5 µmol/L). Effects on glucose homeostasis not seen with equivalent (5 mmol/L) amount of sodium monofluorophosphate (MFP); plasma diffusable fluoride always below 0.04 mg/L (2 µmol/L); protein-bound MFP did not affect glucose homeostasis. |
Rigalli et al. 1995 |
Water intake and fluoride intake approximately 6 times higher in diabetics than in controls for a given fluoride concentration; fluoride absorption about 75% in diabetics versus 63% in controls; fluoride retention about 40% (39-42%) in diabetics versus increasing with fluoride dose (27-45%) in controls. Plasma and tissue fluoride concentrations increased with fluoride dose, significantly higher for diabetics than for controls. Plasma fluoride (mg/L) in controls: 0.008-0.010, 0.015-0.017, 0.029, and 0.072-0.082; in diabetics: 0.0097-0.012, 0.036-0.046, 0.10-0.12, and 0.26-0.36.f Bone fluoride (ppm in ash) in controls: 171-194, 410-560, 872-1,330, and 2,500-3,600; in diabetics: 200-310, 1,000-2,000, 2,700-4,700, and 6,800-9,500. Same mean blood glucose value (453.5 ± 8.2 mg/dL) given for initial and final values in diabetic rats—one of them is probably not correct; for controls, initial value of 121.9 ± 1.7 mg/dL and final value of 129.6 ± 1.7 mg/dL. Markers examined: plasma urea, glucose (nonfasting), creatinine, calcium, phosphorus, uric acid, cholesterol, total protein, albumin, total bilirubin, alkaline phosphatase, glutamate oxaloacetate transaminase; urine urea, creatinine; creatinine clearance; histological evaluations; bone marrow sister chromatid exchanges. Significant differences in many parameters between normal and diabetic animals; with respect to fluoride intake, significant differences only for diabetic rats with fluoride at 50 mg/L (lower plasma cholesterol, higher total protein in plasma, increased width of tibial cortex). |
Dunipace et al. 1996 |
Species and Strain |
Exposure Conditions |
Concentration or Dosea |
Exposure Duration |
Rabbits (Dutch-Belted, female, 3 1/2 months old at start, 1.55 kg) See also Table E-11 |
Drinking water |
0 and 100 mg/L [7-10.5 mg/kg/day]g |
6 months |
Rats (Sprague-Dawley, male, 30-40 g weanlings at start, 432 animals total) |
Drinking water Either calcium-deficient diet or diet deficient in protein, energy, or total nutrients |
5, 15, or 50 mg/L [0.5, 1.5, or 5 mg/kg/day]c |
16 or 48 weeks |
Rats (Charles River, Wistar, females, normal and with streptozotocin-induced diabetes, 8 per group) C: normal, no fluoride in water F10: normal, fluoride in water D: diabetic, no fluoride in water DF10: diabetic, fluoride in water FF: normal, with fluoride intake adjusted to match that of DF10 (1.6-3 mg/day per rat) |
Drinking water and feed (NaF in drinking water) |
Drinking water: Groups C and D, 0 mg/L Groups F10 and DF10, 10 mg/L Group FF, adjusted to match fluoride intake of DF10 Feed: 13 ppm (all groups) [C: 1.0-1.5 mg/kg/day F10: 2.1-2.9 mg/kg/day D: 2.2-2.5 mg/kg/day DF10: 8.4-18.6 mg/kg/day FF: 8.3-11.8 mg/kg/day]i |
3 weeks |
Horses (6 total, thoroughbreds, average age 5 years, average weight 509 kg, euthanized at end of experiment) |
Sevoflurane anesthesia |
Not available |
Mean, 18.5 hours |
Effects |
Reference |
Statistically significant (P < 0.05) increase in serum glucose (17%). Increased IGF-1 (40%). Insulin or other regulators of serum glucose were not measured. No effect of fluoride on serum urea, creatinine, phosphorus, total protein, albumin, or bilirubin; serum glutamate oxaloacetate transaminase; or total alkaline phosphatase. Increased serum fluoride (0.728 versus 0.0441 mg/L)h and bone fluoride (6,650-7,890 versus 850-1,150 ppm in ash). |
Turner et al. 1997 |
No significant effect on fasting plasma glucose; specific data by fluoride treatment group not reported. Combination of general malnutrition and calcium deficiency was not examined. |
Dunipace et al. 1998 |
Normal rats had similar intakes of feed and water regardless of fluoride intake; final body weights were similar. Diabetic rats had 3-5 times higher water intake than normal rats and almost twice the feed intake; final body weights for group D were lower than for normal rats; final body weights for group DF10 were lower than initial body weights. Increase in overall severity of diabetes and higher fasting blood glucose concentrations in fluoride-treated diabetic rats; about 400 mg/dL (22 mM/L) in DF10 versus 250 mg/dL (14 mM/L) in D and 90 mg/dL (5 mmol/L) in C, F10, and FF. Plasma fluoride (approximate, mg/L): C, 0.029; F10, 0.038; D, 0.038; DF10, 0.095; FF, 0.057.j Bone (femoral) fluoride (approximate, ppm in ash): C, 400; F10, 600; D, 400; DF10, 1000; FF, 1900). Fluoride treatment in nondiabetic rats did not cause significant alteration of blood glucose concentrations. |
Boros et al. 1998 |
Mean plasma fluoride after 8 hours was 0.7-0.9 mg/L (38-45 µmol/L). Total and ionized calcium decreased over time; ionized calcium remained within normal limits; total calcium below normal values after 2 hours. Serum glucose concentrations increased throughout, exceeding normal concentrations at 6 hours and thereafter, but within the values commonly observed during general inhalation anesthesia in horses; glucosuria also present after 10 hours. |
Driessen et al. 2002 |
Species and Strain |
Exposure Conditions |
Concentration or Dosea |
Exposure Duration |
Rats (Wistar, adult females, 150-170 g at start; fluoride administered during pregnancy and lactation)k |
NaF orally by feeding tube |
40 mg/kg/day NaF (18 mg/kg/ day fluoride to the mothers) |
Day 6 of gestation through day 21 of lactation |
Rats (Wistar FL, males, 14 weeks old, 8 treated, 10 controls) |
Intraperitoneal injection |
35 mg/kg NaF (15.8 mg/kg fluoride) in physiological saline Controls, saline only |
Single dose, sacrificed 90 minutes later |
aInformation in brackets was calculated from information given in the papers or as otherwise noted. bBased on average of initial and final mean body weights. cBased on water consumption of about 10% of body weight, with no significant differences in body weight with fluoride intake. dBased on water consumption of about 10% of body weight and feed consumption of about 8% of body weight, with no significant differences in body weight with fluoride intake. eBased on final (6-month) mean body weights of 508.8 g for controls and 445.4 g for diabetics, with pretermination (3- and 6-month combined) metabolic data for fluoride intake. fPlasma fluoride (µmol/L) in controls: 0.42-0.54, 0.8-0.9, 1.5, and 3.8-4.3; in diabetics: 0.51-0.65, 1.9-2.4, 5.5-6.1, and 13.6-19.2 gBased on average daily water consumption of 163 mL, mean initial weight of 1.55 kg, and mean final weight of 2.33 kg for the fluoride-treated group. hSerum fluoride: 38.31 versus 2.32 µmol/L. IBased on average daily fluoride intake for days 1-4 with average initial body weight for all groups and average daily intake for days 15-21 with average final body weight for the group. jPlasma fluoride (approximate, µmol/L): C, 1.5; F10, 2; D, 2; DF10, 5; FF, 3. kIn many mammalian species, maternal fluoride exposures are not well reflected by fluoride concentrations in milk; therefore, the impacts of fetal exposure and of reduced milk production by the mothers must also be considered. |
Effects |
Reference |
Marked hypoglycemia in mothers and offspring, attributed to reduced feed consumption. Reduced serum protein content, significant increases in serum sodium and potassium. Significant recovery on withdrawal of NaF or supplementation with vitamins C, D, and E. |
Verma and Guna Sherlin 2002a |
Hyperglycemia (47% increase), accompanied by impairment in renal function, decreased calcium concentrations (13%). |
Grucka-Mamczar et al. 2005 |
TABLE E-17 Effects of Fluoride on Other Endocrine Organs in Humans
Study Population(s) |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
76 male and female inmates of Japanese mental hospital Observational study; summary of cases; cross-sectional |
Thought to be from pesticide use |
Not available Chronic |
41 Russian males with fluorosis, ages 33-45, 19 controls (no contact with fluorine compounds) Case-control study; cross-sectional; full details not available |
Occupational exposure |
Not available >15 years for some |
Volunteers in Argentina, 6 adults Experimental study; subjects included the authors of the report and members of their laboratory |
Oral administration to fasting persons |
27 mg of fluoride (60 mg of NaF) [0.4 mg/kg]b Single dose |
25 young adults (14 males, 11 females) in India with endemic fluorosis (skeletal and enamel), ages 15-30 years (nonobese, nonsmokers, no personal or family history of diabetes mellitus or hypertension) 25 controls with normal fluoride intake (age, sex, and body mass index matched; comparable social and working conditions) Case-control study; cross-sectional for all; longitudinal for subjects initially found to have impaired glucose tolerance; tests were repeated after 6 months on a low-fluoride water source |
Drinking water |
2-13 mg/L in drinking water [0.067-0.43 mg/kg/day]c Controls: < 1 mg/L [< 0.03 mg/kg/day]c Since birth |
Poland, residents of Skawina (living in the vicinity of an aluminum smelter) and Chorzów (employed in any of 3 industries); approximately 50 individuals per group (approximately 200 total) Ecologic measure of exposure (exposure to environmental fluorides from industrial pollution) |
Airborne fluorides Skawina: chronic exposure to fluorine compounds Chorzów: chronic exposure to environmental fluorides and other toxic compounds |
8-10 times the Maximum Allowable Concentration for fluoride of 1.6 µg/m3 (12.8-16 µg/m3) |
Effects |
Reference |
Endocrine disturbances including melanosis in 20 of 76 patients; attributed to dysfunction of parathyroids and adrenals, reversed upon treatment for chronic fluorine poisoning. |
Spira 1962 |
Elevated follicle-stimulating hormone and decreased testosterone in blood in all men with fluorosis; elevated blood luteinizing hormone in men with long-term exposure (>15 years). |
Tokar’ and Savchenko 1977 |
After 1 hour, significant fall of plasma insulin concentrations and increased fluoride; reduced insulin response to glucose challenge. Plasma fluoride: 0.1-0.3 mg/L (5-15 µmol/L). |
Rigalli et al. 1990 |
Impaired glucose tolerance (IGT) in 40% (6 males, 4 females); fasting serum fluoride concentrations positively correlated (P < 0.01) with area under glucose curve in those 10; effect appeared to be reversible on provision of drinking water with “acceptable” fluoride concentrations (<1 mg/L). For all 25 endemic fluorosis patients, significant positive correlation between serum fluoride and fasting serum immunoreactive insulin; significant negative correlation between serum fluoride and fasting glucose:insulin ratio. Normal serum calcium, inorganic phosphorus, and vitamin D; elevated serum alkaline phosphatase in patients with endemic fluorosis. Urine fluoride (mg/L): fluorosis patients, 2-8; controls, 0.2-0.5. Serum fluoride (mg/L): patients with IGT, 0.08 ± 0.04; patients with normal glucose tolerance, 0.02 ± 0.01; controls, 0.01 ± 0.009; IGT patients after 6 months on low-fluoride water, 0.02 ± 0.01. |
Trivedi et al. 1993 |
Excessive excretion of fluorides in urine (53-100% with urine fluoride > 2.3 mg/L; for Skawina, mean = 5.6 mg/L; SD = 2.5, n = 46), associated with a decrease in urine and erythrocyte magnesium concentrations (36-65% with urine magnesium < 5.4 mg/L); increased blood glucose and lactate concentrations, which were normalized by magnesium supplementation. For Skawina, 74% had blood glucose results above the norm (70-100 mg/dL or 3.89-5.55 mmol/L; n = 42). |
Kedryna et al. 1993 |
Study Population(s) |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
U.S., female osteoporosis patients (patients with previous history of hyperparathyroidism and several other conditions were excluded) Initial recruitment included 203 in-state patients from previous fluoride trials and 95 controls who had not taken fluoride; of these, 40 fluoride patients and 43 controls were scheduled for appointments; 15 fluoride patients were no longer taking fluoride or failed the appointments; 5 controls failed the appointments; final study included 25 fluoride patients and 38 controls (mean ages, 70.1 for fluoride group, 69.5 for controls) Cross-sectional study; fluoride-treated patients and non-fluoride-treated controls recruited from database of osteoporosis patients of one investigator; fasting samples; analyses of drinking water, blood, and urine performed blindly; results reported as means of groups and as number outside the normal range for the parameter; urine and plasma fluoride clearly different between groups; no significant difference in mean water fluoride concentrationsSee also Table E-12 |
Slow-release sodium monofluorophosphate plus 1,500 mg/day calcium carbonate Most controls (n = 38) had calcium supplementation |
23 mg/day (mean dose) [0.33 mg/kg/day]b 1.4-12.6 years (mean, 4.2 years) |
China, healthy adults (approximately 120 per group, with either normal or inadequate nutritional intakes; mean ages of groups, 44.9-47.7 years) Cross-sectional cohort study; subjects grouped by location (water fluoride concentration) and nutritional status; populations generally similar (e.g., socially and economically); estimated fluoride intakes and measurements of urine and plasma fluoride and other parameters were made for individuals but results reported only for groups; probably overlap between low (<0.3 mg/L) and middle (around 1 mg/L) fluoride exposure groups for each nutritional category; no mention of whether analyses were performed blindly See also Table E-12 |
Drinking water Normal nutrition defined as > 75 g/day protein and Ca >600 mg/day Inadequate nutrition defined as <60 g/day protein and Ca <400 mg/day |
0.23, 1.02, and 5.03 mg/L (normal nutrition) 0.11, 0.90, and 4.75 mg/L (inadequate nutrition) Estimated intakes: 1.70, 3.49, and 14.8 mg/day (normal nutrition); 1.20, 2.64, 15.32 mg/ day (inadequate nutrition) At least 35 years of continuous residency in the study area |
Effects |
Reference |
Mean fasting blood glucose concentrations 104.7 (SD = 53.0) for fluoride-treated group and 95.2 (SD = 10.3) for controls (difference not considered significant); 3 of 25 fluoride-treated individuals outside normal range (versus 1 of 38 controls). Urine fluoride (mg/L, mean and SD): fluoride group, 9,7 (4.1); controls, 0.8 (0.5); plasma fluoride (mg/L, mean and SD)d: fluoride group, 0.17 (0.068); controls, 0.019 (0.0076). |
Jackson et al. 1994 |
No significant differences in mean blood glucose concentrations among groups. Not clear whether samples were fasting or nonfasting. |
Li et al. 1995 |
Study Population(s) |
Exposure Conditions |
Concentration or Dosea and Exposure Duration |
2 postmenopausal women in Argentina Experimental study; subjects were members of the authors’ department who were receiving NaF as treatment for osteoporosis and who volunteered to undergo glucose tolerance tests; tests were administered in the fasting state |
Treatment for osteoporosis |
13.6 mg/day (30 mg/day NaF) [0.23 mg/kg/day]e 9 and 24 months |
24 women and 2 men, ages 44-66, former residents of an area of endemic fluorosis in Argentina Ecologic exposure measure; cross-sectional study; fasting blood samples |
Drinking water |
Not stated Chronic |
U.S., 199 adult volunteers (mean ages of groups, 62.3, 58.6, 57.2 years) Ecological study; cross-sectional; subjects grouped by location (water fluoride concentration); subjects not randomly selected; nonfasting samples; urine and plasma fluoride concentrations significantly different for groups; study parameters reported by groups; no information on whether analyses were performed blindly See also Table E-12 |
Drinking water, natural fluoride Dietary calcium and calcium concentrations in drinking water were not discussed |
0.2, 1.0, 4.0 mg/L [0.003, 0.01, 0.06 mg/kg/day]b At least 30 years of continuous residency in their communities |
160 males ages 20-50 years, in Mexico Ecologic exposure measure based on occupation; exposure groups overlapped; no information on selection of subjects |
Drinking water alone for 27 men (low group) Occupational exposure and drinking water for 133 men (high group) |
3.0 mg/L in drinking water 2-13 mg/day estimated for low group [0.03-0.19 mg/kg/day]b 3.4-27.4 mg/day estimated for high group [0.05-0.39 mg/kg/day]b Chronic (at least 1 year for occupational exposure) |
aInformation in brackets was calculated from information given in the papers or as otherwise noted. bBased on 70-kg per person. cBased on consumption of 2 L of drinking water per day by a 60-kg adult. dReported as 9.0 (3.6) µmol/L for the fluoride group and 1.0 (0.4) µmol/L for the controls. eBased on 60-kg per person. |
Effects |
Reference |
Disturbed glucose homeostasis when given glucose tolerance test. Plasma F: 0.11 and 0.13 mg/L (5.6 and 6.7 µM/L). |
Rigalli et al. 1995 |
Inverse relationship between plasma fluoride and area under curve of insulin during a standard glucose tolerance test. Plasma F: 0.01-0.18 mg/L (0.5-9.2 µM/L). Urine F: > 1.1 mg/day. |
de la Sota et al. 1997 |
No significant differences among mean glucose concentrations (nonfasting); all mean values were within normal ranges. |
Jackson et al. 1997 |
Elevated follicle stimulating hormone; decreased testosterone, inhibin B, and prolactin; apparent reduction in sensitivity of the hypothalamic-pituitary axis to negative feedback action from inhibin B. Fluoride exposures of the two groups overlapped, and occupational exposures included other chemicals besides fluoride. |
Ortiz-Perez et al. 2003 |