Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
REPRODUCTIVE EFFECTS 4 OF FLUORIDE REPRoDucTnrE EFFECTS IN HUMANS There are no published reports in the literature on reproductive toxicity of fluoride in men. However, two Russian studies showed that chronic occupational exposure to fluoride-contaminated compounds might affect reproductive function. Men who had worked in the cryolite industry for 10-25 years and who demonstrated clinical skeletal fluorosis showed decreases in circulating testosterone and compensatory increases in fol- licle-stimulating hormone when compared with controls Prowar and Savchenko, 1977~. Of the exposed men, those exposed to cryolite for 16-25 years had increased lutein~zing-hormone levels as compared with men exposed for 10-15 years. Women exposed occupationally to air heavily laden with superphosphates demonstrated increases in menstrual irregularities and genital irritations when compared with unexposed controls (Kuznetzova, 1969~. However, occupational exposure to many other compounds in the cryolite and superphosphate industries makes it difficult to implicate any one substance, such as fluoride, in inducing these health effects. A recent study of women employed in silicon wafer manufacturing (fabrication room workers) showed a relative risk of spontaneous abor- tions of 1.45 times that of women (of the same ages) who worked in nonfabrication rooms (Schenker et al., 1992~. The overall increase in 73
74 Health Effects of Ingested Fluoride risk ranged from about 20% to 40%. There was a dose-response rela- tionship and a consistency of findings for persons exposed to one specific class of solvents. Spontaneous abortions were also associated with fluoride exposure but only in one work group, and a strong dose-response was not present. The authors characterized the fluoride-associated increase in relative risk of spontaneous abortions as "less consistent" than the results of exposure to some solvents in this stucly and "less consis- tent" with other research. REPRODUCTIVE EFFECTS IN AwIMA~s A summary of the reproductive effects of fluoride in animals is pre- sented in Table 4-~. Several of the earliest studies of the effects in rodents noted an adverse effect when the diet contained fluoride at more than 100 mg/kg. Phillips et al. (1933) concluded that the upper limit of safety of chronic intake of fluoride by the rat in terms of effects on estrus cycle, rate of reproduction, and lactation was 20 mg/kg of body weight per day. In their study, fluoride was added to the diet as sodium fluoride (190 mg/kg) or rock phosphate (sodium fluoride at 210 and 350 mg/kg). Ale fluoride concentration of the stock diet, which also contained bone meal, was not stated. Most "natural ingredient" rodent diets containing bone meal that are available today, however, have fluoride concentrations ranging from 10 to 50 mg/kg. The bioavailability of fluoride from these diets is 45-50% (Whitford, 1991~. In their study of the effects of fluoride on the reproductive perfor- mance of male rats, Araibi et al. (1989) added fluoride at 0, 100, or 200 m~/k~ to the "standard" rat diet. . .~, . ~ -~ ~ The fluoride concentration of the standard diet was not reported. After 60 days, blood was collected from some of the rats for determination of testosterone concentration, and the testes were prepared for microscopic examination. The fertility of the remaining rats was tested by mating them with normal females for 4 days. The number of pregnancies and offspring was reduced significantly in the 200-mg/kg group but not in the lOO-mg/kg group. The litter sizes did not differ among any of the groups. The authors commented} that the treated rats showed "less interest toward females." However, food and water intakes, body-weight changes, ant! activity levels were not reported and were probably reduced in the 200-mg/kg group. If so, those effects
Reproductive Effects of Fluoride 75 might have been involved in the apparent loss of interest. The diameters of the seminiferous tubules were reduced slightly in both fluoride-treateci groups. The thickness of the peritubular membranes was increased, the percentage of tubules with spermatozoa was reduced, and the serum testosterone level was lower in the 200-mg/kg group. Those effects did not occur in the lOO-mg/kg group. Messer et al. (1973) fed mice a low-fluoride diet (0. I-0.3 mg/kg) and drinking water containing fluoride at 0, 50, 100, or 200 mg/L for 33 weeks. The two highest amounts resulted in retarded growth ant] im- paired reproduction; increased mortality occurred in the 200-mg/L group only. No litters were produced by mice in the 200-mg/L group, and only nine litters were born to the 50 mice in the lOO-mg/L group. Six of the nine litters were stillborn or eaten at birth. Litter production in the 50- mg/L group was normal. The fluoride concentrations in ashed humeri were about 100 ppm in the control group and 7,800 ppm in the 50-mg/L~ group. According to Messer et al., the group that received no fluoride showed signs of "fluorine deficiency with a progressive development of infertility in two successive generations." That finding was based on the development of anemia and the fact that fewer litters were producer} in the control group than in the 50-mg/LJ group. The authors concluded that their findings supported the necessity of fluoride in the diet of mice. Tao and Suttie (1976) repeated the study of Messer et al. (1973) as closely as possible, except that the diet contained adequate levels of iron ant! copper and fluoride was adcted to the diet at concentrations of 2 and 100 mg/kg. The stock diet had a fluoride concentration of less than 0.5 mg/kg. Femur-ash fluoride concentrations of the third-generation mice in the 0-, 2-, ant! lOO-mg/kg groups at 33 weeks were 90, 328, and 10,020 ppm, respectively. The investigators noted no differences among the groups with respect to hematocrit or reproduction. They concluded that the fluoricle-deficiency findings of Messer et al. (1973) were probab- ly due to an iron or copper deficiency. The role of fluoride was con- siderec] secondary in that it preventer] fluoride deficiency from develop- ing, apparently by increasing the absorption of iron and copper, both of which were present in marginal concentrations in Messer's mouse diet. Studies of reproduction in cattle have generally shown no effect of fluoricleunIess intake is suff~cientto produce skeletal fluorosis and other adverse effects. The reports by Phillips et al. (1934) and Mitchell and Ed man (1952) indicated that reproductive performance of cattle with
76 Cal .t ¢ - 4 - ~: .c .e o a V] .~ o Cal .= ¢ ~ ~ o Go', o Ct .c o o .a~ ~ ~ .,- o o CO 3 so .O Hi. ~1J g3 ~2° '8 ~ ~E 00 E ~ 64 ; ~ e · ~,°, ~ 5 ~r o 0 ° ~Z 8no 8~ ~o ~ In -._ E ~ ~ ce Z ct Cal ,sb40 O - ] O ~ ~ s-4 ~- 0 ~0 ~0 ~ O ~ ~ be lo - 8 o be be o o ~ o ~ C ~. o' .~ o ~ C~ = C,,.O `: ~ ~ ~ ~ ~ o ~ ~ ~ .5 8 oo -C ~ ~ ~ ~ o^ ~ ~ ° ~ ~o p: p ~o > 5 ~-
4 - o o 4 - o A o -I - ~ v) to - 8 - A o ~ 77 CO oo 4 - Do ~ Cal Pa - ~ oN ·E ~ o .~ e ° 2 o Hi- ~ ~ it_ =~£~= I - be - To .~ - = . . _ Ct ~ Ct o - - ·u 4 - C) ~in. ,D ~ ct g · CO it: o . Cd O ~·t .s O ~ ·- cd O ~D I: O O . ~ _ ~-e,4 ~ ~ _ :, ~ O 3 ~ u~ bO Cr~ ~ ·= .,= to ~ o ~y ~D o ~ Cd ^°° 8 't' - ~ -E ~ 5 ~ E ~ t - ° ~ ~ 8 ·s ~ _ ~_ ~ ~ ~ ~ o~ o oo \° ~ ~ I.,, Oo ° c _ ~o o - - o o bl) o ~ .O U. - ct ._ - . ~0 ~ =m 8 ~ ~ ~ ~ ~o tt ~o ~o ~ ~ ~ 2C 8 8 ~ =.~5 c C. Ct ~n - Cd V) b' c: o .~ Ct 8 .c o o m O ~ _ o C~ 4 - o ~7 00 - bC 00 bB bC o o bO 0 ~O ~a
78 Health Effects of Ingested Fluoride signs of chronic fluorosis was unaffected except for delays in estrus after parturition. Suttie et al. (1972) exposed Holstein heifers to different patterns of dietary fluoride for 6 years. The fluoride doses ranged from that contained only in the forage and grain to supplements of 3.0 mg/kg per day for 4 months, alternating with 0.75 mg/kg per day for ~ months. On an annual basis, the doses given to We fluoride-treated groups were 0.75 and I.5 mg/kg per day. At the end of the study, the fluoride concentration in vertebral ash was 552 ppm in the control group, and the concentrations ranged from 7,660 to 15,200 ppm in the treatment croups. Skeletal fluorosis was evident in all treatment groups. The authors concluded that no differences existed among the groups in growth rate or reproductive performance, although "the conception rate for the entire experiment was low." Milk production was "somewhat less" in animals receiving I.5 mg/kg per day each year, "but as small numbers of animals were involved, the observations are inconclusive." In their study, van Rensburg and de Vos (1966) noted the effects of fluoride on reproductive performance of Afrikaner heifers receiving fluoride at 5-12 mg/L of drinking water. Five groups were given drink- ing water containing fluoride at 5 or ~ mg/L or fluoride at 5, 8, or 12 mg/L and superphosphate (chiefly tribasic calcium phosphate produced by treating phosphate rock with sulfuric acid), which had been treated to reduce its fluoride content. The fluoride concentrations in food, super- phosphate, or drinking water after addition of superphosphate were not stated. No group was given nonfluoridated water, and fluoride con- centrations in bone or other tissues were not determined. Breeding was started 9 months after beginning exposure to fluoride and observations on reproductive performance were made over four breeding seasons. The heifers were bred naturally during the first two breeding seasons and by artificial insemination during the last two seasons. Reproduction did not differ among the groups during the first season. In the second season, anestrus after parturition increased in the 8- ant] 12-mg/L groups. In Me third season, fertility clearly declined in those two groups but was more pronounced in the superphosphate groups. In the fourth season, the effects were even greater as judged by the calving rates and the services per conception in all groups, especially in the X- and 12-mg/L groups that were given superphosphate. Contrary to what was expected, the addition of superphosphate to the water aggravated the effects of fluoricie, which suggested to the authors that the removal of fluoride might have been
Reproductive Effects of Fluoride 79 incomplete and that actual fluoride intakes were more than had been intendecI. Despite those uncertainties, it was concluded that "for normal reproduction the fluorine content of the drinking water should be under 5 my/." Eckerlin et al. (1986a) reported on a problem with milk production on a dairy farm. From January 1977 through August 197S, the average annual milk yield was 7,851 kg per cow per year. The fluoride con- centration of the diet used during that period was not reported. In September 1978, a new commercial concentrate and mineral mix was introduced as a supplement to the diet. During the next 6 years, prob- lems developed in the general health, reproduction, and milk production of the cows. The average fluoride concentration of three bags of the concentrate saved from 1982 was 97 mg/kg. One sample of the mineral mix from 1982 had a fluoride content of 1,415 mg/kg. Milk production during 1978-1984 averaged 5,434 kg per cow per year, a (decline of 3 ~ % . When the investigators conciucted their study at the farm in 1984, the last ossified coccygeal vertebrae of 15 cows were surgically removed and analyzed for fluoride. The fluoride concentrations in bone ash ranged from 300 ppm in a 2-year-old cow to 2,100 ppm in a 6-year-old cow. Considering the adverse effects that were recorded and the reportedly high concentrations of fluoride in the diet, those values were very low. The low fluoride concentrations in bone do not support the contention that the effects were due to fluoride toxicity. As noted above, Suttie et al. (1972) found bone fluoride concentrations of over 15,000 ppm and essentially none of the toxic manifestations reporter] by Eckerlin et al. (1986a). Other studies of the effects of fluoride on reproduction in owls, hens, kestrels, dogs, and mink have been reported. Hoffman et al. (1985) ant! Pattee et al. (198X) provided a diet supplemented with fluoride at 0, 40, or 200 mg/kg to breeding pairs of screech owls (60 and 33 pairs, respec- tively) for 5-6 months. The control diet h act a fluoride concentration of 27.2 mg/kg. In the Hoffman et al. study, no differences were found in the number of eggs laid by all the groups. Egg volume was 6% lower in two of the treatment groups. Five hatchlings weighed slightly less in the 40-mg/kg group and 9% weighed slightly less in the 200-mg/kg group than in the control group, but this difference disappeared 7 days later. Pattee et al. (1988) reported that nesting chronology was similar among all groups with respect to initiation of egg laying, initiation of incubation,
80 Health Effects of Ingested Fluoride hatching date, intervals between any of those events, clutch size, percent fertility, and percent of eggs hatched. The number of young produced per clutch was lower in the 200-mg/kg group than in the control group (2.6 + 0.4 vs. 4.2 + 0.4~. Congenital anomalies were not observed in either study, nor were the concentrations of fluoride in bone. However, Pattee et al. (1988) reported Me fluoride concentrations in eggshell ash. The eggshell concentrations of the control, 40-, and 200-mg/kg groups were 6.4 ppm, 53 ppm, and 87 ppm, respectively. It was of interest that the differences in the 40- and 200-mg/kg groups were not statistically significant. The authors concluded that slight-to-moderate reproduction disorders in screech owls (slightly lower egg volume and initial body weight) might occur in areas heavily polluted with fluoride. Guenter (1979) fed chickens diets containing fluoride at 0-1,300 mg/kg. The results suggested that dietary fluoride at 200 mg/kg per day for ~12 days improved egg production and feed efficiency, although shell thickness was reduced slightly. Dietary fluoride had no effect on fertility or hatchability of chicken eggs. In a second experiment, dietary fluoride at 100 mg/kg improved feed efficiency, shell quality, and the number of collectible eggs. Egg size was reduced only in the groups fed fluoride at 1,000 mg/kg and 1,300 mg/kg. C~cified-tissue fluoride concentra- tions were not reported. The effects on reproduction in 24 pairs of American kestrels fed cockerels with a low or high fluoride content for 10 days were reported by Carriere et se. (1987~. The cockerels fed to the control group had a background femur fluoride concentration of 62 ppm and those fed to the treatment groups had concentrations of 4,513 ppm and 7,691 ppm. Femur fluoride concentrations of the kestrels were proportional to fluo- ride intake. The clutch sizes tended to be smaller as the concentration of fluoride intake increased, but the differences were not statistically sig- nificant. Percent fertility and percent hatchability were not related to fluoride intake. Paste! mink were fed diets supplemented with fluoride at 0, 33, 60, 108, 194, or 350 mg/kg per day for 382 days (Aulerich et al., 1987~. The fluoride concentration of the nonsupplemented diet was 35 mg/kg. There were no important differences among the groups in breeding, gestation, whelping, or lactation. The survivability of kits whelped by minks fed fluoride at 350 mg/kg was markedly decreased. The body weights of those whelped by minks in the 60- and lO8-mg/kg groups
Reproductive Effects of Fluoride 81 were higher than those of the control group at 3 and 6 weeks. The femur fluoride concentrations of the six groups were directly related to the fluoride dose and ranged from 1,362 to 13,706 ppm. Eckerlin et al. (1986b) reported a problem with milk production and survival rates of kits at three fox farms. The farms had successfully produced pelts for several years. In 1985, the foxes experienced severe early postpartum agalactia, which resulted in high mortality of kits during the first few postpartum days. Analysis of the commercial diets used on the three farms revealed fluoride concentrations of 98-137 mg/kg. Control diets from another supplier had fluoride concentrations of 23 and 31 mg/kg. Bone fluoride concentrations increased with age from about 500-900 ppm in the kits to 2,000-3,000 ppm in the adults. The authors concluded that the agalactia was caused by excessive fluoride intake. Reduced milk production associated with high fluoride intake has also been reported for cows (Maylin and Krook, 1982), although, as reported above, Suttie et al. (1972) reported only a marginal effect in cows that were exposed to high fluoride doses. In a followup study at the fox farm, Eckerlin et al. (1988) provided additional evidence that the problems were related to high fluoride concentrations in the diets. The fluoride concentrations of new diets used from 1985 to 19X7 ranged from ~ to 23 mg/kg. Bone fluoride concentra- tions declined, kit production increased anti kit ~'rviv~hilitv ~nnrn~rh~A normal expectations. ~ ~--~ I- ~_^ ~ ~ ~ ~V^~J ~_~ From 1970 to 1980, an increased incidence of perinatal deaths and congenital deformities occurred in a kennel of Shetland sheepdogs (Schellenberg et al., 1990~. The problems started shortly after the kennel was moved from an old frame building to a new concrete-block building. In 1979 and 1980, mottled teeth in the few surviving pups and cranial exostoses in the adults were noted for the first time. These calcified- tissue effects were attributed to the high fluoride content of the diet (460 mg/kg added in the form of rock phosphate). In 19X2, a 2-year study of 20 shelties of proven fertility was started to determine the reproductive effects of high-fluoride diet and kenned well water (Schellenberg et al., 1990~. The well water was of interest after it became clear that some unknown factor in the kennel environment was responsible for the adverse reproductive effects. There were four treat- ment groups (four females and one male in each group) that were given a high- or low-fluoride diet (55 mg/kg) and kennel well water or distilled
82 Health Effects of Ingested Fluoride water. Except for the group fed the low-fluoride diet anti well water (which produced the expected number of pups), the number of pups whelped in the other groups paralleled the poor reproduction results that prompted the study. The two high-fluoride groups produced only two litters and IS pups, 10 (67%) of which were dead within 7 days. The two low-fluoride groups produced six litters and 33 pups, 14 (42%) of which were dead within the first 7 postpartum days. The two well-water groups produced six litters and 39 pups, IS (46%) of which died within 7 days. The two distilled-water groups produced two litters and 9 pups, 5 (56%) of which diec! within 7 clays. It was determined that the perina- tal deaths were not due to agalactia. Overall, the misses! pregnancy rate was 44% and the perinatal death rate was 50%. No serious congenital defects occurred in the high-fluoride groups ant! two occurred in the low-fluoride groups. Several minor congenital defects occurred! in all groups. The authors concluded that, "Although 460 mg/kg F in the dog food did produce bony exostoses, we did not find convincing evidence that it adversely affected reproduction in shelties. The cause of the reproductive problems was apparently not the dog food, water, foliage, genetic factors, or infectious disease ant! currently remains uncleter- mined." The problem was further examiner] in a study using rats (Marks et al., 1984~. Four treatment groups (9 males and lS females in each group) were given the same diet of dog food containing fluoride at 460 or 55 mg/kg and the same sources of drinking water as described above. The rats were brought into the kennel at 39 days of age, given the specified diets for the next 60 days, and then mated. The authors concluded that "even after two litters, the only adverse effect was dental fluorosis in the high-fluoride groups. The results indicated that rats cannot be used in the search for the causers) of reproductive problems in dogs in this kennel." DISCUSSION Adverse effects on reproductive performance associated with high concentrations of fluoride intake have been demonstrated in mice, rats, cattle, owls, hens, kestrels, dogs, mink, and foxes (reducer] lactation). The water or food threshold fluoride concentration associated with these effects in mice, rats, caKle, ant! foxes is approximately 100 mg/L (100 mg/kg). An exception to this was reported by van Rensburg ant] cle Vos
Reproductive Effects of Fluoride 83 (1966) who concluded that the water fluoride concentration should be less than 5 mg/L for normal reproduction in caKle. The authors noted, however, that the fluoride concentration in the superphosphate added to three of the five diets was not known, so the actual concentrations of fluoride in those diets might have been higher than intended. The thresh- old concentration for mink, owls, and kestrels is 100-200 mg/kg and for hens over 500 mg/kg in diet. These dietary fluoride concentrations are much higher than those in the fluoridated drinking water of humans. This fact is consistent with the lack of evidence suggesting a link between consumption of fluoridated water and problems with human reproduction. It was noted in several of the studies outlined above that the fluoride concentrations in the diets of the control groups were unknown or con- siderably higher than is optimal. In these cases, the investigators might not have been able to detect the effects, if any, of fluoride intake at the low ens! of the dosage scale. It is recommended that future studies of reproductive effects of fluoride intake include control diets with fluoride concentrations of less than I.0 mg/kg. To confirm the level of fluoride exposure, it is also recommended that bone fluoride concentrations be measured using accurate ant! reliable preparative and analytical methods, such as the hexamethyIdisiloxane-facilitated diffusion method] of Taves (1968) as modified by Whitford (1989) and the fluoride ion-specific electrode.