Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Toxicological information

Toxicity to reproduction

Currently viewing:

Administrative data

Endpoint:
reproductive toxicity, other
Remarks:
various studies
Type of information:
other: evidence based on degradation product
Adequacy of study:
weight of evidence
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
data from handbook or collection of data

Data source

Referenceopen allclose all

Reference Type:
secondary source
Title:
EU RAR Hydrogen Fluoride
Author:
Various
Year:
2001
Bibliographic source:
EU RAR Hydrogen Fluoride, Volume 8, 2001
Reference Type:
publication
Title:
Multigenerational evaluation of sodium fluoride in rats
Author:
Collins TFX, Sprando RL, Black TN,Shackleford Me, Bryant MA, Olejnik N, Ames MJ, Rorie JI
Year:
2001
Bibliographic source:
Fd. Chem. Toxic, 39, 601-6013
Reference Type:
publication
Title:
Developmental toxicity of sodium fluoride measured during multiple generations
Author:
Collins TFX, Sprando RL, Black TN,Shackleford Me, Bryant MA, Olejnik N, Ames MJ, Rorie JI
Year:
2001
Bibliographic source:
Fd. Chem. Toxic., 39, 867-876.

Materials and methods

Test guideline
Qualifier:
equivalent or similar to guideline
Guideline:
OECD Guideline 416 (Two-Generation Reproduction Toxicity Study)
Principles of method if other than guideline:
The EU RAR on hydrogen fluoride summarizes the results of nummerous studies (attached in Section 13). The study performed by the US-FDA were pubilshed in the two papers by Collin et at. cited here as reference due to their significance for the overall conclusion and derivation of an NOAEL.
GLP compliance:
not specified

Test material

Constituent 1
Chemical structure
Reference substance name:
Zinc hexafluorosilicate
EC Number:
240-894-1
EC Name:
Zinc hexafluorosilicate
Cas Number:
16871-71-9
Molecular formula:
F6Si.Zn
IUPAC Name:
zinc(2+) hexafluorosilanediuide

Test animals

Species:
rat
Strain:
other: CD CRL:CD-BR
Sex:
male/female

Administration / exposure

Route of administration:
oral: drinking water
Details on mating procedure:
P animals were treated for 10 weeks before mating and subsequently mated (1:1) within each group until either pregnancy was determined by the presence of a sperm plug or for three consecutive weeks. After the mating P males were transferred to a follow up study for effects on the male reproductive system (see Sprando et al. (1997, 1998)). At day 20 of gestation 8 P females of each group were sacrificed. These dams and their offspring were examined in a separate study (Collins et al. 2001b). The remaining dams were allowed to litter. On postnatal day 4 the litters were culled to 10 pups per group by random procedure. The F1 animals remained within their respective treatment groups and at postnatal day 21, 36 F1 animals from each sex were randomly selected for further breeding. These F1 animals were kept for another 10 weeks and subsequently mated according to the same procedure as the P animals. At gestation day 20, F1 females were sacrificed and dams and their fertility and offspring was examined.
Duration of treatment / exposure:
Twenty-two days old P0 males and females (Sprague-Dawley rats 12/sex/group) received fluoride for about 14 weeks (10 weeks pre-mating, 3 weeks mating and 1 week post-mating periods) and pregnant P0-females continued to be exposed until the end of lactation. The F1 generation remained within the same treatment groups as their parents. F1 animals received F- in utero, via lactation and via the drinking water to approximately 14 weeks after weaning.
Frequency of treatment:
daily, drinking water
Doses / concentrationsopen allclose all
Dose / conc.:
0 mg/L drinking water
Dose / conc.:
25 mg/L drinking water
Dose / conc.:
100 mg/L drinking water
Dose / conc.:
175 mg/L drinking water
Dose / conc.:
250 mg/L drinking water
No. of animals per sex per dose:
48

Examinations

Parental animals: Observations and examinations:
In the
male rats the following parameters were monitored (both P0 and F1 generations): testes weight
and histology (Leydig cell morphology, composition of interstitial cell population, seminiferous
tubule morphology, Sertoli cell morphology, spermatid development), homogenisation-resistant
spermatid counts, sperm cell production and production rate (per gram of testis tissue),
secondary sex organ weights, serum LH, FSH and testosterone concentrations, body, liver,
spleen, heart and adrenal weights. Only isolated statistical differences in various organ weights
between treated and control groups or between P0 and F1 generations were observed. These
differences were not treatment related and do not indicate toxicologically relevant effects.

Results and discussion

Results: P0 (first parental generation)

General toxicity (P0)

Clinical signs:
effects observed, non-treatment-related
Body weight and weight changes:
effects observed, treatment-related
Description (incidence and severity):
P females and males showed a negative dose-related trend in body weight gain, but only in the males of the 250 mg/l group the reduction in body weight gain became statistically significant.
Food consumption and compound intake (if feeding study):
effects observed, treatment-related
Description (incidence and severity):
In the F1 females a negative dose-related trend in feed consumption was seen, but
statistical significance was not reached. In the same period, P and F1 males and females
consumed less drinking water than the control animals, possibly due to decreased drinking water
palatability.

Reproductive function / performance (P0)

Reproductive performance:
no effects observed
Description (incidence and severity):
No effects were observed on any of the following parameters indicative for reproductive
performance or pup viability: mating index, fertility index, time to mating, and for the F1
generation number of implants, litter size, number of stillborn pups per litter, and pup survival
and lactation indices. Runts were randomly distributed among control and treatment groups.

Details on results (P0)

In a two-generation study in rats NaF was administered via the drinking water.at concentrations
of 0 (<0.2), 25, 100, 175 and 250 mg NaF/l (Sprando et al. , 1997). Based on the reported
average body weights of the adult P0 males and an assumed daily water intake of 50 ml/d these
concentrations are equivalent to approximately 0, 1.1, 4.4, 7.5 and 10 mg F-/kg b.w./d,
respectively. The low fluoride diet contained 7.95 mg F-/kg (equivalent to 0.4 mg/kg b.w./d).
Twenty-two days old P0 males and females (Sprague-Dawley rats 12/sex/group) received
fluoride for about 14 weeks (10 weeks pre-mating, 3 weeks mating and 1 week post-mating
periods) and pregnant P0-females continued to be exposed until the end of lactation. The F1
generation remained within the same treatment groups as their parents. F1 animals received F- in
utero, via lactation and via the drinking water to approximately 14 weeks after weaning. In the
male rats the following parameters were monitored (both P0 and F1 generations): testes weight
and histology (Leydig cell morphology, composition of interstitial cell population, seminiferous
tubule morphology, Sertoli cell morphology, spermatid development), homogenisation-resistant
spermatid counts, sperm cell production and production rate (per gram of testis tissue),
secondary sex organ weights, serum LH, FSH and testosterone concentrations, body, liver,
spleen, heart and adrenal weights. Only isolated statistical differences in various organ weights
between treated and control groups or between P0 and F1 generations were observed. These
differences were not treatment related and do not indicate toxicologically relevant effects.
Testicular histological examination did not reveal any abnormalities. The NOAEL for effects on
male reproductive organs and cells in this study is 250 mg NaF/l which is equivalent to
approximately about 10 mg F-/kg b.w./d.
In an additional paper (Sprando et al, 1998) the authors described the results of an
electronmicroscopical morphometric analysis of testicular tissue of F1 generation males
(5/group). The animals were taken from the same experiment as described above.
Seminiferous tubules comprised 89%, 87%, 88%, 88% and 88% of the total testis volume while
the interstitial space occupied 9.3%, 11.2%, 10.2%, 9.8% and 9.9% of the total testis volume for
the 0, 25, 100, 175 and 250 mg NaF/l treatment groups, respectively. Statistically significant
differences between control and NaF-treated rats were not observed with respect to absolute
volume of the seminiferous tubules, interstitial space, Leydig cells, blood vessels boundary layer,
lymphatic space, macrophages, tubular lumen or absolute tubular length and absolute tubular
surface area, mean Sertoli cell nucleoli number per tubular cross-section, mean seminiferous
tubule diameter and the mean height of the seminiferous epithelium. A statistically significant
decrease in the absolute volume and volume percent of the lymphatic endothelium was observed in
the 175 and 250 ppm NaF-treated groups and in the testicular capsule in the 100 ppm NaF-treated
groups. The significance of this finding is unknown at the present time. Overall, the quantitative
information obtained suggests that exposure to NaF at the doses used in the present study does
not adversely affect testis structure or spermatogenesis in the rat.

Effect levels (P0)

Key result
Dose descriptor:
NOAEL
Effect level:
250 mg/L drinking water
Based on:
not specified
Sex:
male
Basis for effect level:
body weight and weight gain
reproductive function (sperm measures)

Results: F1 generation

General toxicity (F1)

Clinical signs:
effects observed, non-treatment-related
Description (incidence and severity):
A two-generation study according to a randomized design, with mating protocol and treatment in
compliance with the OECD-416 guideline, performed under GLP, has been reported by Collins
et al. (2001a). For the P generation, 48 rats per sex (strain: CD CRL:CD-BR) were assigned to
groups receiving 0 (<0.2 mg F-/l), 25, 100, 175 or 250 mg NaF/l via the drinking water. The
animals were kept on a low fluoride diet containing 7.95 mg F-/kg.
P animals were treated for 10 weeks before mating and subsequently mated (1:1) within each
group until either pregnancy was determined by the presence of a sperm plug or for three
consecutive weeks. After the mating P males were transferred to a follow up study for effects on
the male reproductive system (see Sprando et al. (1997, 1998)). At day 20 of gestation 8 P
females of each group were sacrificed. These dams and their offspring were examined in a
separate study (Collins et al. 2001b). The remaining dams were allowed to litter. On postnatal
day 4 the litters were culled to 10 pups per group by random procedure. The F1 animals
remained within their respective treatment groups and at postnatal day 21, 36 F1 animals from
each sex were randomly selected for further breeding. These F1 animals were kept for another 10
weeks and subsequently mated according to the same procedure as the P animals. At gestation
day 20, F1 females were sacrificed and dams and their fertility and offspring was examined.
The following observations were made:
- feed and drinking water consumption, clinical observations, and growth and development
in and F1 animals,
- pathology: 10 males and females from each group (P generation, F1 weanlings, F1 adults)
gross lesions, body weights, organ weights of 14 tissues (among which male sex organs
and ovaries), histopathology of 41 tissues in all animals and in addition histopathology of
17 extra tissues in the animals of the control and highest dose groups,
- reproductive performance: date of birth, litter size, mating-, gestation-, fertility- and
viability-indices pup survival. Indicators for embryo/foetal toxicity and effects on foetal
development (both skeletal and visceral) were reported separately (Collins et al. 2001b).
No dose-related clinical effects were observed. During the 10-week periods before mating, P
males in the highest dose group consumed statistically significant less feed than the control P
males. In the F1 females a negative dose-related trend in feed consumption was seen, but
statistical significance was not reached. In the same period, P and F1 males and females
consumed less drinking water than the control animals, possibly due to decreased drinking water
palatability. P females and males showed a negative dose-related trend in body weight gain, but
only in the males of the 250 mg/l group the reduction in body weight gain became statistically
significant. No effects on body weights were observed in the F1 animals. Absolute and relative
organ weights of any of the organs studied were not affected.
Body weight and weight changes:
no effects observed
Description (incidence and severity):
No effects on body weights were observed in the F1 animals.
Food consumption and compound intake (if feeding study):
effects observed, treatment-related
Description (incidence and severity):
In the F1 females a negative dose-related trend in feed consumption was seen, but
statistical significance was not reached. In the same period, P and F1 males and females
consumed less drinking water than the control animals, possibly due to decreased drinking water
palatability.
Organ weight findings including organ / body weight ratios:
no effects observed
Description (incidence and severity):
Absolute and relative organ weights of any of the organs studied were not affected.
Other effects:
effects observed, treatment-related
Description (incidence and severity):
F1 males and F1 females showed dose-related and statistically significant mild dental whitening
at 100, 175 and 250 mg NaF/l but not at 25 mg NaF/l. Dental staining or mottling was not
observed. In the 250 mg/l dose group, in all adult males and females of the P and F1 generations
and in 8 female and 10 male F1 weanlings, an increase in the development of prominent growth
lines (basophilic lines in dentin and dental enamel) was observed. In the same exposure group,
dentin deposition in de pulpal cavity was observed in one P male and one P female and in four
F1 males. Hyperkeratosis of the limiting ridge of the forestomach was observed in one P male
and three P females of 100 mg/l group, in one P male of the 175 mg/l group, in one P male of the
250 mg/l group and in two F1 males and two F1 females of the 250 mg/l group.

Effect levels (F1)

Key result
Dose descriptor:
NOAEL
Generation:
F1
Effect level:
250 mg/L drinking water
Based on:
not specified
Sex:
male/female
Basis for effect level:
body weight and weight gain

Overall reproductive toxicity

Key result
Reproductive effects observed:
no
Lowest effective dose / conc.:
11.7 mg/kg bw/day
Treatment related:
yes

Any other information on results incl. tables

No dose-related clinical effects were observed. During the 10-week periods before mating, P males in the highest dose group consumed statistically significant less feed than the control P males. In the F1 females a negative dose-related trend in feed consumption was seen, but statistical significance was not reached. In the same period, P and F1 males and females consumed less drinking water than the control animals, possibly due to decreased drinking water palatability. P females and males showed a negative dose-related trend in body weight gain, but only in the males of the 250 mg/l group the reduction in body weight gain became statistically significant. No effects on body weights were observed in the F1 animals. Absolute and relative organ weights of any of the organs studied were not affected.

F1 males and F1 females showed dose-related and statistically significant mild dental whitening at 100, 175 and 250 mg NaF/l but not at 25 mg NaF/l. Dental staining or mottling was not observed. In the 250 mg/l dose group, in all adult males and females of the P and F1 generations and in 8 female and 10 male F1 weanlings, an increase in the development of prominent growth lines (basophilic lines in dentin and dental enamel) was observed. In the same exposure group, dentin deposition in de pulpal cavity was observed in one P male and one P female and in four F1 males. Hyperkeratosis of the limiting ridge of the forestomach was observed in one P male and three P females of 100 mg/l group, in one P male of the 175 mg/l group, in one P male of the 250 mg/l group and in two F1 males and two F1 females of the 250 mg/l group. No effects were observed on any of the following parameters indicative for reproductive performance or pup viability: mating index, fertility index, time to mating, and for the F1

generation number of implants, litter size, number of stillborn pups per litter, and pup survival and lactation indices. Runts were randomly distributed among control and treatment groups. For the detection of developmental toxicity 8 pregnant P dams per group were sacrificed and their offspring was examined for gross appearance (Collins et al. 2001b). For each treatment group, 29 to 34 pregnant F1 dams were sacrificed and their offspring was studied extensively for skeletal and visceral abnormalities. Neither in the 8 P dams nor in the F1 dams clinical signs of toxicity were observed during

gestation. In the P dams and in the F1 dams at 100 mg/l and above a reduced average intake of drinking water was observed, which reached statistical significance in the P dams at 250 mg/l and in the F1 dams at 175 and 250 mg/l. Although in all treatment groups in both generations a reduced feed intake was observed, this effect showed no dose-response relationship. No dose response relationship was observed in body weight gain during pregnancy or in gravid uterine weight. In both P and F1 generations, no effects were observed on female fertility parameters, or on parameters indicative for embryo- or foetotoxicity. Foetal body weights and lengths were not affected. No gross external malformations in the offspring of both generations were observed. Upon examination for skeletal defects, in the offspring of the F1 dams no effects were observed with respect to sternebral variations. An increased incidence of retarded ossification of the hyoid bone in the 250 mg/l group was observed when the data were analysed on a total number of pups basis, but not on a per litter basis. In none of the other skeletal elements any indication for retarded ossification due to treatment with NaF was observed. No irreversible structural changes, neither in the skeleton nor in the soft tissues were found.

From the study (Collins et al. 2001a, 2001b) it can be concluded that no fertility effects occurred in rats exposed to NaF in the drinking water at concentrations up to 250 mg/l. The dose levels (expressed on a mg/kg b.w./d basis) for P and F1 animals were about equal, and based on drinking water consumption data and body weights, the exposure level of 250 mg/l is on average equal to 10.7 mg F-/kg b.w./d or 12.5 mg F-/kg b.w./d for males and non-pregnant females respectively. The NOAEL for developmental toxicity in this study is 250 mg NaF/l drinking water, which is equal to 11.7 mg F-/kg b.w./d for the P generation and equal to 12.7 mg/l for the F1 generation. (dose level representative for the period of gestation).

Embryo-, feto- and developmental toxicity

As part of a more extended study into the kinetics and toxicity of fluoride administered as various salts, pregnant rats received 0, 51, 124 and 200 mg NaF/kg diet (equivalent to 0, 1.13, 2.74 or 4.41 mg F-/kg b.w./d) from day 1 to day 20 of gestation. Observations with respect to reproduction included number of viable fetuses/litter, fetal and placental weight and resorptions. No effects were observed. This study is incomplete with respect to reproductive toxicity of fluoride because the offspring was not appropriately examined (Theuner et al 1971). In another study, pregnant female rats (20/group) were exposed to 0 or 30 mg F-/l as NaF (~ 0 or 3.36 mg F-/kg b.w./d) through day 6-15 of gestation and killed at day 20. It was stated that in a preliminary study with 30, 40 and 50 mg F-/l no maternal toxicity was observed at 30 mg/l. Observations included maternal weight, verification of fetal life, sex, macroscopic deformities, resorptions, corpora lutea, fetal histopathology and bone malformations. Enhanced numbers of congenital deformities (equimosis, bone formation deficiencies of sternum and skull bones) and runts were observed. The histological examination were reported to be normal, but specification about techniques and tissues were not given. The LOAEL in this study is 30 mg F-/l (~ 3.36 mg/kg b.w./d. (Larez et al 1980). The study is inadequately reported and maternal toxicity was tested in an indirect way (no effect on maternal weight change in a pilot study). The relevance of the effect denoted by the term "equimosis" cannot be assessed. Collins et al. (1995) have investigated the oral developmental toxicity of NaF in rats in an assay similar to OECD-414. The substance was administered via the drinking water (ultra pure) in concentrations of 0, 10, 25, 100, 175 and 250 mg NaF/l, daily throughout gestation. Groups of 33 to 37 females were mated with untreated males. Actual dose levels were 0, 0.63, 1.76, 7.06, 11.12 and 11.35 mg F-/kg b.w. All animals received a low fluoride diet containing 7.95 mg F-/kg feed which results in an additional exposure to approximately 0.6 F- mg/kg b.w./d. The observations included maternal toxicity (behavioral and clinical signs including dental

mottling, feed and drinking water intake, body weight) and toxicity in the off-spring (numbers of live and dead foetuses, implantations, resorptions, numbers of corpora lutea, sex, weight, external examination, skeletal abnormalities and soft tissue abberations). Clinical signs of toxicity in the dams were not seen, but at the highest two dose levels drinking water consumption and at the highest dose level feed intake were diminished. At 250 mg/l dams showed reduced growth. At the highest level of exposure, a slight reduction in corpora lutea, and number of implants per dam were seen but these observations were no effect of fluoride. No signs of retarded foetal development were obtained. At the highest dose level a limited but statistically significant increase in the number of foetuses with skeletal variations was found. The number of litters affected was not significantly increased. Thus this study did not reveal relevant reproductive or developmental toxicity resulting from fluoride at dose levels up to 250 mg/l. At

this level maternal toxicity was observed. The NOAEL for maternal toxicity and developmental effects in this study is 11.12 mg F-/kg b.w./d. (Collins et al. 1995).

In another study by NTP, similar to OECD-414, sodium fluoride was administered via the drinking water to rabbits and rats.

Groups of 26 female rats received 0 (0.6), 50, 150 or 300 mg NaF/l in deionised water from day 6 through day 15 of gestation. The feed contained about 12.4 mg/kg fluoride (average value).

Actual intake of fluoride via the drinking water amounted to 0.3, 3.0, 8.4 and 12.3 mg F-/kg b.w./d. Fluoride in feed resulted in an additional exposure of 1.0 mg F-/kg b.w./d. With the highest dose administered the authors aimed at the induction of some maternal toxicity while avoiding dehydration effects due to reduced water intake resulting from bad palatability of NaF solutions. Observations included maternal toxicity (clinical signs, weight gain, water and feed consumption, liver and kidney weights) and toxicity in the offspring (number of litters, corpora lutea, implantation sites, resorptions, late death, live fetuses, fetal body weights, sex ratio, external and skeletal and soft tissue malformations.

The only significant effect on the dams was a reduced water intake in the high dose group during the treatment. No indications for embryo-, feto- or developmental toxicity were obtained. The NOAEL in this study for developmental of fetotoxicity is therefore equal to the highest level tested being 12.3 mg/kg b.w./d (Heindel et al. 1996).

In the same study report (Heindel et al. 1996), the results of exposure of groups of 26 rabbits to 0 (0.6), 100, 200 and 400 mg NaF/l in the drinking water for day 6 through 19 of gestation have been described. Exposure via the drinking water amounted to 0.1, 4.7, 8.2 and 13.2 mg F-/kg b.w./d, whereas exposure via food contributed about 0.8 mg F-/kg b.w./d to the total fluoride intake. As with the rats observations included maternal toxicity (clinical signs, weight gain, water and feed consumption, liver and kidney weights) and toxicity in the offspring (number of litters, corpora lutea, implantation sites, resorptions, late death, live fetuses, fetal body weights, sex ratio, external and skeletal and soft tissue malformations. Dams exposed to the highest fluoride level demonstrated reduced water intake during the exposure period and a reduced feed intake on days 6 through 8 of gestation resulting in a reversible loss of body weight over these days. No (other) signs of maternal toxicity were obtained. Examination of uteri and offspring did not reveal any sign of embryo-, feto- or developmental toxicity. The NOAEL for reproductive effects in this study is therefore 13.2 mg/kg b.w./d (highest level tested).

In a preliminary study, only reported as an abstract, I.P. injection of 0 or 15 mg NaF/kg b.w. in rats (~ 0 or 6.79 mg F-/kg b.w.) through days 7-14 or days 14-20 of gestation resulted in skeletal malformations and retardations in sternebrae and thoracic vertebrae with increased occurrence of 14th rib. These aberrations were mainly observed in the 14-20 day dose group. Maternal toxicity was not observed. For a proper evaluation, the whole report should be made available. (Horvath 1989).

NaF was intraperitoneally or subcutaneously injected into pregnant rats from days 10 through 18 of gestation. A more or less dose-related increase in the frequency of necrotic placentas and dead fetuses was observed. Even the lowest dose level (1 mg NaF/kg b.w./d; equal to 0.45 mg F-/kg b.w./d) elicited these effects, which were more pronounced after IP than after SC administration of NaF. Up to 9 mg F-/kg b.w./d. no maxillo-facial malformations were seen. Other effects were not studied. (Deveto et al. 1972).

From the two generation studies by Collins et al. (2001a, 2001b; see section generation studies) a NOAEL for developmental toxicity of 250 mg NaF/l drinking water can be derived (highest dose tested).

A study into the impact of fluoride in drinking water on human fertility showed a decrease in total fertility rate (determined as number of birth per 1000 women) associated with increasing fluoride drinking water concentrations (Freni 1994). However, this epidemiological study was performed at population level, and a causal relationship between fluoride exposure and reduced fertility rate was not demonstrated. Too many confounding factors, which were not taken into account, may be involved in the association between fluoride intake and the biological phenomenon. This study is unsuitable to derive a NOAEL.

Applicant's summary and conclusion

Conclusions:
The available information from repeated dose toxicity studies as well as reproductive toxicity studies do not indicate a hazard for female reproduction in relation to exposure to fluoride. A recent two-generation study focussing on the damage to the male reproductive system, no testicular effects were seen at an exposure level of 250 mg NaF/L. Additionally, it is especially noted that in none of the animal carcinogenicity studies effects on morphology ofany of the reproductive organs were found.
Executive summary:

No specific data with respect to the reproductive, developmental or embryo/feto-toxicity of HF have been reported. However, these effects are systemic and thus, because of similar kinetics and dynamics, data on sodium fluoride are used to give insight in the reproductive toxicity of HF.

Human data on reproductive toxicity of fluoride are inconclusive.

The available information from repeated dose toxicity studies as well as reproductive toxicity studies do not indicate a hazard for female reproduction in relation to exposure to fluoride. In several studies indications were obtained that oral exposure to fluoride may damage testicular tissue and reduce male fertility. The LOAEL for these effects was 2.26 mg F-/kg b.w./d. In a study in which F- (as NaF) was administered directly into the testes of rats no substance related effects were seen. In another recently completed two-generation study of high quality (US-FDA) which specifically focussed on damage to the male reproductive system, no testicular effects were seen at an exposure level of 250 mg NaF/l (equivalent to about 10 mg F-/kg b.w./d).

The available results of the long-term studies, the test for the effects of F- on testes after intratesticular injection and the reports of the two-generation study by US-FDA contradict the results of the tests which indicate that F- may damage testes tissue.

For the establishment of an overall NOAEL/LOAEL the following considerations are taken into account:

- In all animal studies in which fertility effects were reported, NOAELs could not be found. It cannot be excluded that this is the result of some unknown contaminant.

- In these studies dose levels are generally below back-ground exposure, while concentrations in food or drinking water have not explicitly been determined. Therefore the relevance of the findings is highly questionable.

- These studies are very limited in design and technical realisation and of considerable less quality than the studies performed by US-FDA, their weight of evidence is therefore far less. In contrast, the US-FDA studies are technically well performed according to modern standards, while high maximum dose levels were used, with well-described background fluoride exposure.

- In a dermal irritation study (Derelanko et al. 1985) a reduced testes weight (with 40%) was found after 4 hours of exposure to 2% HF (~40 mg HF/kg b.w.) under occlusion (observation time: 96 h after exposure) without significant microscopic alterations, however, dermal lesions were already observed at 0.01 % HF (0.2 mg/kg b.w.).

- In the 90-days inhalatory toxicity test with HF in rats no specific effects on reproductive organs were seen up to the highest level of exposure (7.52 mg/m3) which roughly corresponds to a systemic daily dose of 0.856 mg F-/kg b.w.

- In the NaF drinking water carcinogenicity studies by NTP (1990) with rats and mice weights of reproductive organs were not determined. However, at histopathology no effects on testes or ovaries were observed in these studies. In the NaF diet study on carcinogenicity with mice (Maurer et al. 1993) no changes in reproductive organ weights were seen; histopathology data were not available. In an identical study with rats (Maurer et al. 1990) neither weight changes nor histological changes of the reproductive organs were seen. (see 4.1.2.8). In none of these studies it was mentioned that sperm characteristics were evaluated.

Consequently, the NOAEL from the US-FDA two-generation study (250 mg NaF/l, equivalent toabout 10 mg F-/kg b.w./d) is used as a NOAEL for effects on fertility.

In the available studies on embryo- and developmental toxicity, some embryotoxicity was observed. From three studies similar to OECD-414, for NaF a NOAEL of 11.12 mg F-/kg b.w./d for maternal toxicity and developmental effects is established, which was further confirmed in a two-generation study in compliance with the OECD-416 guideline. In this two generation study the observed maternal toxicity, if any, was limited to mild dental whitening, and changes in dentin and enamel structure, next to some hyperkeratosis in the upper gastrointestinal tract.