Domain-specific effects of prenatal fluoride exposure on child IQ at 4, 5, and 6–12 years in the ELEMENT cohort

https://doi.org/10.1016/j.envres.2022.112993Get rights and content

Highlights

  • The effect of prenatal fluoride exposure on IQ is stable over time.

  • The effect of prenatal fluoride exposure on IQ is domain specific.

  • Prenatal fluoride exposure is associated with deficits in performance IQ.

  • Prenatal fluoride exposure is marginally associated with deficits in verbal IQ.

Abstract

Objective

Prenatal exposure to fluoride has been associated with adverse neurodevelopmental outcomes. However, the neuropsychological profile of fluoride's developmental neurotoxicity at low levels and the stability of this relationship across childhood has not been characterized. We investigated the longitudinal and domain specific effect of prenatal fluoride exposure on IQ among children ages 4, 5, and 6–12 years in the Early Life Exposures in Mexico to Environmental Toxicants (ELEMENT) cohort.

Methods

We measured the average of maternal urinary fluoride at each trimester of pregnancy adjusted for creatinine (MUFCRE). Children were administered the McCarthy Scales of Children's Abilities at ages 4 (N = 386) and 5 (N = 308), and the Wechsler Abbreviated Scale of Intelligence at age 6–12 (N = 278). We used generalized estimating equation (GEE) models to estimate the population averaged effect of MUFCRE concentration on longitudinal General Cognitive Index (GCI)/Full-Scale IQ (FSIQ), Verbal IQ (VIQ), and Performance IQ (PIQ) scores (N = 348). We tested for possible interactions between MUFCRE and child sex as well as for MUFCRE and time point on children's IQ. All models controlled for relevant available covariates.

Results

The mean/median MUFCRE concentration was 0.90/0.83 mg/L (SD = 0.39; IQR, 0.64–1.11 mg/L). A 0.5 mg/L increase in MUFCRE predicted an average 2.12-point decrease in GCI/FSIQ (95% CI: −3.49, −0.75) and 2.63-point decrease in PIQ (95% CI: −3.87, −1.40). MUFCRE was marginally associated with VIQ across time (B = −1.29, 95% CI: −2.60, 0.01). No interactions between MUFCRE and child sex or MUFCRE and time were observed.

Conclusion

The negative association between prenatal fluoride exposure and longitudinal IQ was driven by decrements in non-verbal intelligence (i.e. PIQ), suggesting that visual-spatial and perceptual reasoning abilities may be more impacted by prenatal fluoride exposure as compared to verbal abilities.

Introduction

Fluoride is added to drinking water and salt for the prevention of dental caries (CDC, 2016). Other sources of fluoride include fluoridated dental products and supplements, certain foods that absorb naturally occurring fluoride, such as green and black tea, and foods that are sprayed with fluoride-containing pesticides (i.e., grapes; Nutrient Data Laboratory, 2015; Zohoori et al., 2013). Recent studies conducted in the United States (Abduweli Uyghurturk et al., 2020), Canada (Till et al., 2018), and Mexico (Thomas et al., 2016; Castiblanco-Rubio et al., 2021) have reported positive associations between fluoride from dietary sources, including drinking water and salt, and urinary fluoride levels in pregnant women. Because of its ubiquity and its ability to pass through the placenta and blood-brain barrier to reach the fetal brain (Agency for Toxic Substances and Disease Registry, 2003), the safety of fluoride exposure in pregnancy has received much attention, both in endemic fluorosis areas (Jiménez et al., 2017) and communities that have fluoridation programs (Green et al., 2019; Bashash et al., 2017).

While it is not disputed that fluoride is a developmental neurotoxicant at high exposure levels, there are relatively few studies that have assessed fluoride's potential neurotoxicity at levels found in fluoridated areas (i.e., 0.7 mg/L), particularly for pregnant women and young infants. In 2021, the National Toxicology Program (National Toxicology Program, 2020) conducted a systematic review on the impact of fluoride on neurodevelopmental outcomes. The NTP identified two high-quality prospective birth cohort studies that were conducted in Mexico City where salt is fluoridated at 250 ppm (Bashash et al., 2017) and in Canada where drinking water is fluoridated at 0.7 mg/L (Green et al., 2019). Both cohort studies found a 4-to 6-point lower Full-Scale IQ score in children per 1 mg/L increase in maternal urinary fluoride level; in the Canadian cohort the effect of maternal urinary fluoride on IQ was only found in boys while the effect of drinking water fluoride level on IQ was found in both boys and girls.

Most human developmental toxicology studies focus on global or composite test scores, such as Full-Scale IQ, which are derived from a diverse set of tasks (Kamphaus, 2019). While global outcomes are considered highly significant from a public health (Lanphear, 2015) and economic standpoint (Gould, 2009), a low composite score does not convey specific information about the child's intellectual and cognitive profile. When there are strengths and weaknesses in a cognitive profile, the use of a composite score may have high sensitivity, but at the cost of low specificity (Fiorello et al., 2007). Partitioning Full-Scale IQ into domain specific intellectual abilities, such as verbal and nonverbal skills, may reveal particular cognitive domains that are more sensitive to neurotoxic exposures or may be differentially affected over time (Bellinger et al., 2016). For example, studies have demonstrated that prenatal and early-life lead exposure is more strongly associated with non-verbal intelligence compared to verbal intelligence between the ages of 2–7 (Bellinger et al., 1991; Desrochers-Couture et al., 2018; Dietrich et al., 1991, 1993; Factor-Litvak et al., 1999; Jusko et al., 2008; Wasserman et al., 1997). Similarly, prenatal and early-life fluoride exposure has been associated with greater deficits in non-verbal abilities than verbal abilities in preschool years (Till et al., 2020; Farmus et al., 2021; Cantoral et al., 2021); however, other studies examining early-life exposure to fluoride did not find observe this profile (Ibarluzea et al., 2021) or did not report verbal and non-verbal intelligence (Bashash et al., 2017). Whether non-verbal intelligence is associated with early life exposure to fluoride over the course of child development has not been examined.

In the present study, we examined the longitudinal and domain specific effects (i.e., verbal and nonverbal intelligence) of prenatal fluoride exposure on IQ in mother-child dyads from the Early Life Exposures in Mexico to Environmental Toxicants (ELEMENT) cohort. Since our prior publication on this cohort (Bashash et al., 2017), we received additional maternal urinary fluoride and creatinine data enabling us to examine children's IQ at three separate time points (age 4, 5, and 6–12 years) and using a larger sample size at each time point relative to our prior work. We also examined the potential for sex-specific effects based on findings that boys may be more susceptible to prenatal exposure than girls (Cantoral et al., 2021; Comfort and Re, 2017; Green et al., 2019, 2020; Torres-Rojas and Jones, 2018).

Section snippets

Study sample

The ELEMENT project enrolled mother-child pairs from three hospitals in Mexico City serving low to middle income families. Participants were recruited as part of four longitudinal birth cohort studies and are described in a recent cohort profile paper (see Perng et al., 2019). Of the four cohorts, cohorts 2 A and 3 had prenatal information and archived maternal urine samples collected during pregnancy. Cohort 2 A included 327 women recruited between 1997 and 1999 for an observational study of

Sample characteristics

Out of the 997 mothers from cohorts 2 A and 3, 971 were at least 18 years of age. Of the 971, 585 had urinary fluoride and urinary creatinine measured for at least one trimester (6 were excluded for having MUFCRE concentrations greater than 3.5 standard deviations from the mean). Of the 585 mothers with MUFCRE data, 391 had MSCA data at age 4, 314 had MSCA data at age 5, and 282 had WASI data at age 6–12. The final sample included 348 mother-child dyads with complete covariate, MUFCRE, and

Discussion

We examined the association between prenatal fluoride exposure and IQ scores in children of mothers included in the Early Life Exposures in Mexico to Environmental Toxicants (ELEMENT) study. Consistent with past research conducted on a smaller number of mother-child dyads from this cohort (Bashash et al., 2017), an increment of 0.5 mg/L in maternal urinary fluoride concentration was associated with a 2-point decrement in children's Full-Scale IQ scores. Results remained consistent when we

Limitations

A limitation of our study is that fluoride exposure was estimated through maternal non-fasting spot urine samples. Urinary fluoride has a short half-life (approximately 5 h), and measurement of fluoride may be diluted by lack of control for behaviours that could contribute to acute changes in fluoride levels, such as consumption of fluoride-free bottled water prior to urine sampling. We minimized these limitations by collecting serial urine samples across more than one trimester of pregnancy

Conclusion

In conclusion, prenatal exposure to fluoride is associated with sustained impacts on IQ. Non-verbal abilities may be more susceptible to impairment from prenatal fluoride exposure as compared to verbal abilities. These results were found among mother-child pairs living in a region of Mexico in which fluoride is added to salt. These findings contribute to the growing body of evidence on fluoride's neurotoxicity, and indicate a need to develop recommendations for pregnant women. Future research

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors gratefully acknowledge: The American British Cowdray Hospital for providing facilities for the ELEMENT research and Dr. David Bellinger for his collaboration on the design and execution of the ELEMENT study's cognitive testing.

Funding: This data was supported by the U.S. National Institutes of Health (NIH; grants R01ES021446 and R01-ES007821); the National Institute of Environmental Health Sciences/the U.S. Environmental Protection Agency (NIEHS/EPA; grant P01ES022844, 83543601),

References (66)

  • R. Niu et al.

    Effects of fluoride on synapse morphology and myelin damage in mouse hippocampus

    Chemosphere

    (2018)
  • D.B. Thomas et al.

    Urinary and plasma fluoride levels in pregnant women from Mexico City

    Environ. Res.

    (2016)
  • C. Till et al.

    Fluoride exposure from infant formula and child IQ in a Canadian birth cohort

    Environ. Int.

    (2020)
  • D. Abduweli Uyghurturk et al.

    Maternal and fetal exposures to fluoride during mid-gestation among pregnant women in northern California

    Environ. Health

    (2020)
  • Toxicological Profile for Fluorides, Hydrogen Fluoride, and Fluorine (Update)

    (2003)
  • M. Arora et al.

    Teeth as a biomarker of past chemical exposure

    Curr. Opin. Pediatr.

    (2013)
  • M. Bashash et al.

    Prenatal fluoride exposure and cognitive outcomes in children at 4 and 6–12 years of age in Mexico

    Environ. Health Perspect.

    (2017)
  • D. Bellinger et al.

    Low-level lead exposure and children's cognitive function in the preschool years

    Pediatrics

    (1991)
  • S. Bobek et al.

    Effect of long-term fluoride administration on thyroid hormones level blood in rats

    Endocrinol. Exp.

    (1976)
  • B.M. Caldwell et al.

    Home Observation for Measurement of the Environment

    (1984)
  • G.A. Castiblanco-Rubio et al.

    Dietary influences on urinary fluoride over the course of pregnancy and at one-year postpartum

    Biol. Trace Elem. Res.

    (2021)
  • Water Fluoridation Data & Statistics

    (2016)
  • N.C. Chaitanya et al.

    A systematic analysis on possibility of water fluoridation causing hypothyroidism

    Indian J. Dent. Res.

    (2018)
  • A. Cinar et al.

    Effects of chronic fluorosis on thyroxine, triiodothyronine, and protein-bound iodine in cows

    Fluoride

    (2005)
  • N. Comfort et al.

    Sex-specific neurotoxic effects of organophosphate pesticides across the life course

    Curr. Environ. Health Rep.

    (2017)
  • A.S. Ettinger et al.

    Effect of calcium supplementation on blood lead levels in pregnancy: a randomized placebo-controlled trial

    Environ. Health Perspect.

    (2009)
  • P. Factor-Litvak et al.

    The Yugoslavia Prospective Study of environmental lead exposure

    Environ. Health Perspect.

    (1999)
  • C.A. Fiorello et al.

    Interpreting intelligence test results for children with disabilities: is global intelligence relevant?

    Appl. Neuropsychol.

    (2007)
  • A. Gomaa et al.

    Maternal bone lead as an independent risk factor for fetal neurotoxicity: a prospective study

    Pediatrics

    (2002)
  • R. Green et al.

    Association between maternal fluoride exposure during pregnancy and IQ scores in offspring in Canada

    JAMA Pediatr.

    (2019)
  • R. Green et al.

    Sex-specific neurotoxic effects of early-life exposure to fluoride: a review of the epidemiologic and animal literature

    Curr. Epidemiol. Rep.

    (2020)
  • E. Gould

    Childhood lead poisoning: conservative estimates of the social and economic benefits of lead hazard control

    Environ. Health Perspect.

    (2009)
  • J.E. Haddow et al.

    Maternal thyroid deficiency during pregnancy and subsequent neuropsychological development of the child

    N. Engl. J. Med.

    (1999)
  • Cited by (13)

    View all citing articles on Scopus
    View full text