x0000x00001 Community Dentistry Oral Epidemiology Fluoride intake and - PDF document

1 ��1 Community Dentistry Or
��1 Community Dentistry Oral Epidemiology Fluoride intake and urinary fluoride excretion in 4- and 8-year-old c�ildren living in urban d rural areas of Sout�west Nigeria. Aut�ors: O. Ibiyemi, F.V. =o�oori,�.A Valentine, A. �aguireCentre for Oral Healt� �esearc�, Sc�ool of Dental Sciences, Newcastle University, UK nstitute of Healt� and Society, Newcastle University, UK Sc�ool of Healt� and Social Care, Teesside University, UK �unning titleFluoride intake and UFE in 4- and 8-year-oldeywords: Fluoride, intake, excretion, retention, Nigeria, c�ildren Corresponding aut�or: Professor Anne �aguire Centre for Oral Healt� �esearc�, Sc�ool of Dental Sciences Newcastle University, Framlington Place Newcastle upon Tyne, NE2 4BW il: anne.maguire@ncl.ac.uk Tel: �44 �0) 1912088564 Fax: �44 �0) 1912085928 ��2 Abstract Objectives: To estimate and compare total daily fluoride intake �TDFI), daily urinary fluoride excretion �DUFE), daily fluoride retention �DF�), fractional urinary fluoride excretion �FUFE) and actional fluoride retention �FF�) in 4- and 8 year-old Nigerians and explore associations between outcomes to improve understanding of fluoride metabolism.. et�ods: Using a cross-sectional observational study, 72 four year-olds and 72 eig�t year-olds re recruited from nursery and primary sc�ools �respectively) in lower and �ig�er water F areas of n and rural localities in Oyo State, sout�-west Nigeria. TDFI from diet and toot�paste ingestion was assessed using a validated Food Frequency Questionnaire and visual scale of toot�paste used toot�brus�ing. DUFE was measured by collecting 24-�our urine sample, FUFE estimated as t�e ratio between DUFE and TDFI, DF� estimated as TDFI-TDFE �w�ere TDFE = DUFE � stimated faecal F excretion �i.FI x10%), and FF� was estimated >�TDFI-DF�)/TDFI] x Data were analysed using ANOVA wit� post-�oc tests and Students t tests and strengt�s of ssociations between k

2 ey variables measured. sults�
ey variables measured. sults�ean �SD) TDFI, DUFE, DF�, FUFE and FF� were 0.137 �0.169) mg/kg bw/d, 0.032 7) mg/kg bw/d, 0.091 �0.147) mg/kg bw/d, 44% �44%) and 46% �44%), respectively for 4-ear-olds. Corresponding values for 8-year-olds �n=63) were 0.106 �0.130) mg/kg bw/d, 0.022 7) mg/kg bw/d, 0.073 �0.107) mg/kg bw/d, 36% �30%) and 54% �30%), respectively. Dietary ontribution to TDFI was 79% and 75% �respectively), for 4- and 8-year-olds. �ean �SD) TDFI om toot�paste ingestion was 0.021 �0.013) mg/kg bw/d in 4-year-olds,0.014 �0.010) mg/kg bw/d in 8-year-olds �p=0.002) but wit� no differences between areas. Differences in dietary F intake termin t�e main differences in F exposure between areas. T�e positive correlation between FI and DUFE was weak for 4 year-olds �r = �0.29) and strong for 8 year olds �r = �0.64). A positive correlation was observed between TDFI and DF� for bot� age groups: �r) = �0.98 for 4-year-olds and �r) = �0.99 for 8-year-olds. sion: Fluoride intake in t�e 4- and 8 year-old Nigerians was muc� �ig�er t�an t�e optimal range´ of 0.05-0.07 mg/kg bwin rural, �ig�er F water areas,wit� diet as t�e main contributor. F retention was similar in bot� age groups, wit� almost �alf of TDFI retained in t�e y. In terms of risk versus benefit for fluorosis and dental caries, t�is finding s�ould be onsidered w�en mitigating against excessive fluoride exposure and planning F-based prevention. ��3 Introductioncessive systemic exposure to fluoride �F) during toot� development can increase t�e risk of veloping dental fluorosis . C�ronic excessive intake of F from birt� to four years of age is considered to be a major contributor to t�e development of dental fluorosis in aest�etically important teet�. Excessive F exposure in older c�ildren can affect posterior teet�, alt�oug� it is important to avoid too precise definitions of ages of greatest risk of

3 fluorosis since t�ere is incre
fluorosis since t�ere is increasing evidence t�at developing toot� germs may be vulnerable to excessive F exposure over an even longer period . In view of t�e risk periods for dental fluorosis, it is important to understand and aim to control sources of excessive systemic F intake especially during periods of enamel formation. A total daily F intake �TDFI) of 0.05-0.07 mg of F per kg of body weig�t �bw) �as been considered as �optimal´ during teet� mineralization to provide greatest resistance to dental aries and minimal risk of dental fluorosis. T�is guidance, alt�oug� empirically-derived, �as been adopted by many countries . Information on TDFI is important for public �ealt� planners and �ealt� care professionals w�en planning community-based F t�erapy, as well as w�en mitigating against potential over-exposure to F . Variations in F intakes across different communities arise from differences in dietary and toot�brus�ing �abits as well as al environmental conditions, suc� as source of water and its F concentration, temperature and consequently t�e volume of fluid intake. In addition, F intake wit�in and between communities can vary considerably depending on individual¶s diet as well as erences in t�e F concentration of t�e community¶s water supply used to prepare and cook meals. T�e risk of dental fluorosis is positively correlated wit� body F burden or F retention, a value � can be derived subtracting values for xcretion from F intake . Factors suc� as ype of diet �eg. meat-based or vegetarian), urinary flow rate and renal tubular fluid pH can impact F retention by increasing or reducing t�e absorption and excretion of ingested F to t�e practical difficulties associated wit� collecting accurate data on F exposure in �ildren, 24� urinary F excretion �as been considered as a good biomarker of total daily F intake . A compre�ensive study using data for 212 c�ildren younger t�an 7 years, w�o consumed esternised diet¶, concluded t�at reasonably good estimations of total F intake and retention can be obtained from daily urinary F excretion data at community levels H

4 owever, t�e latter study sugges
owever, t�e latter study suggested t�at furt�er studies of F intake and excretion in c�ildren ��4 older t�an 7 years are needed to explore �ow F intake and excretion data are related in older c�ildren. In addition, due to t�e diversity of diet among western- and eastern-based societies, studies from developing countries would �elp to improve understanding of t�e lations�ip between type of diet �e.g. a vegetarian-based diet) and F excretion and retention. T�ere are few data from sub-Sa�aran Africa, wit� t�e majority of researc� in t�ese locations being observational studies of dental fluorosis 9- and skeletal fluorosis , 13. In order to identify t�e primary sources of exposure to F and more fully understand t�e local risk factors for fluorosis, wit� a view to t�eir mitigation in t�e longer-term, t�is study was designed to estimate TDFI, DUFE, TDF�, and t�e strengt� of t�e associations between t�ese variables in nd 8 year-old Nigerian c�ildren living in naturally fluoridated areas. et�ods T�e study was undertaken according to t�e guidelines in t�e Declaration of Helsinki ; t�e study protocol was approved by t�e Et�ics Committees of Faculty of �edical Sciences, Newcastle University, UK and t�e University of Ibadan/University College Hospital, Nigeria. T�is study represented P�ase 2 of a larger study in w�ic� t�e prevalence of dental defects of namel in Nigerian four and eig�t-year-old c�ildren was determined and was undertaken tween t�e end of t�e dry season �February) and mid-rainy season �July) in 2013. Two Local Government Areas �LGAs), Ibadan Nort� and Ibarapa Central, wit� a population size of approximately 306,795 and 102,979 respectively, were randomly selected from a list of all 33 LGAs in Oyo State �population size: 5.592 million), Nigeria 16. Fluoride analysis of 124 water samples collected from common community ground water supplies �wells and �oles) in rural and urban locations in t�ese two LGAs was undertaken to identify 4 areas, mely

5 urban �ig�er .80 ± 1
urban �ig�er .80 ± 1.0 ppm), rural �ig�er F �2.0 ± 3.0 ppm), urban lowe - nd rural lower F �0.06 - 0.07ppm) and t�ese formed t�e setting in w�ic� t�e study s subsequently undertaken. T�e cluster sampling of 624 four �n=302) and eig�t �n=322) ear-olds of bot� genders undertaken in randomly selected nursery and primary sc�ools in P�ase 1 of t�e overall project was based on a power of 95% at an alp�a level of 5% to termine a difference in mout� prevalence of DDE of 3% between areas wit� a non-completion rate of 30%. A 23% subsample �to allow for attrition) of t�ese c�ildren was ndomly selected and consented to participate in t�is present study. Inclusion criteria ��5 included residence in t�e study area since birt�, being �ealt�y, wit� no �istory of metabolic se or acid-base disturbance and not receiving a t�erapeutic diet. Heig�t �m) was measured wit�out s�oes using a portable stadiometer �DE56618903; ADE Germany) and weig�t �kg) was measured wit�out �eavy clot�es and s�oes using a portable digital scale �SOEHNLE, Slim Design Linea, Germany). Participants¶ parents/guardians were interviewed about t�eir c�ild/ward¶s dietary and toot�brus�ing �abits recorded using a previously validated, standardised, interviewer-dministered semi-quantitative food frequency questionnaire �FFQ) and a pictorial scale of mount of toot�paste routinely used 18. FFQs are used to measure food and drink intakes duals over a specific period of time, depending on t�e aim of t�e study �eg. over t�e previous 3-6 mont�s, year or longer). T�e FFQ used was developed to include locally onsumed diets and toot�pastes and translated into local language ��oruba), after w�ic� it was pre-tested among mot�ers wit� similar socio-demograp�ic c�aracteristics as t�e mot�ers of study participants and local language modified to ensure its reliability and validity. Samples of �ome-made

6 food and drink consumed by t�e
food and drink consumed by t�e c�ildren plus drinking and cooking waters were collected from �ouse�olds and ready--consume samples of foods and drinks purc�ased from local s�ops as appropriate. A private interview was used to clarify t�e nature of food /drinks and obtain recipes if necessary. All food and drink samples collected were individually �omogenized and stored frozen at -C at t�e University of Ibadan, Nigeria before being transported on dry ice to Newcastle University, UK for appropriate F analysis. From a list of all toot�pastes used by study participants, t�e same brands of toot�pastes were purc�ased locally and stored at room temperature prior to F analysis. Food and drink samples were pooled, based on t�eir food/drink category, prior to F analysis. Waters and non--based drinks were measured directly using a F-Ion-Selective-Electrode F-ISE) after addition of TISAB III, w�ile a �examet�yldisiloxane-facilitated diffusion s used to measure t�e F concentration of food, milk-based drink and toot�paste samples 19. A UK-based F database was used to provide a best estimate of F content of any ssing drink and food group samples since it reports F concentrations for similar foods and drinks cooked/prepared wit� similar water F concentrations. Eac� c�ild¶s daily dietary F ake �DDFI), according to drink and food category, was estimated by multiplying t�e F oncentration �mg/kg) of eac� category by t�e amount �kg) consumed per day and t�en ��6 summing F intakes from eac� food and drink group to derive an estimate of F intake on a y weig�t basis �mg/kg bw/d). amount of F ingestion �/d) from toot�paste was estimated multiplying t�e pictorially recorded amount of toot�paste used per brus�ing �by its F concentration /kg) and recorded frequency of daily use. T�is value was t�en derived on a bodyweig�t sis �as mg/kg bw/d) and multiplied by 41%: t�e mean % of toot�paste ingested per �brus�ing session reported for UK 4 to 6 year-olds and Iranian 4-year-olds 22since t�ere is scarcity of global data for 8-year-olds and lack of any data from Nigeria. Total

7 daily F intake �TDFI), in mg/ds
daily F intake �TDFI), in mg/ds t�en estimated by adding DDFI and daily F intake from toot�brus�ing and t�eir relative contributions determined. TDFI on y weig�t basis /kg bw/d), for eac� individual c�ild, was t�en calculated by dividing TDFI by body weig�t. stimate daily urinary F excretion �DUFE), in mg/d, a 24-�our urine sample was collected e from eac� study participant using t�e met�od described by =o�oori and �ugg-Gunn, its volume �) recorded and its F concentration �µg/sured using a F-ISE and direct 19. Daily urinary F excretion was t�en estimated by multiplying urine volume and F oncentration and t�e value normalised for individual bodyweig�t g Total daily F excretion �TDFE) t�roug� urine and faeces was estimated by assuming t�at a fraction of 10% of TDFI is excreted t�roug� faeces and adding t�is to t�e DUFE 23, 24FE value was t�en used to estimate total daily F retention �DF�) �g bw/d) using t�e following formula: DF� = TDFI-TDFE. In addition, to explore t�e body t�res�olds for xcretion and retention of F in t�ese 2 age groups, estimates were derived for fractional urinary F excretion �FUFE) and fractional F retention �FF�), based on t�e following equations: FUFE �%) = >DUFE/TDFI] x100 �%) = >TDF�/TDFI] x 100 completeness of 24-�our urine samples was assessed by comparing urinary flow rate �ml/�our) wit� t�e World Healt� Organization �WHO) reference ranges for 4- 160 ml/�our) and 8- year-olds �9-300ml/�our) . Participants wit� a urine flow rate outside t�is range were excluded from furt�er analysis. T�e reliability and reproducibility of t�e F ��7 analytical met�ods was examined by re-analysing 10% of samples and to confirm t�e validity analytical met�od, a known concentration F standard was added to anot�er 10% of t�e samples, prior to re-analysis to measure F recovery. Descriptive analysis was undertaken using SPSS version 21 �

8 ;SPSS, C�icago, IL, USA) to der
;SPSS, C�icago, IL, USA) to derive mean �SD) values for eac� group. Statistically signi¿cant dirences among areas were detected using ANOVA t�en investigated using a post-�oc test �Tukey) wit� statistical ni¿cance set at Į < 0.05. A studenttest was used to compare t�e key variables between t�e two age groups. Pearson correlation was used to measure t�e strengt� of t�e associations tween TDFI and; �i) DUFE; �ii) DF�, andiii) FF�. Based on Fis�er's r-transformation 25, t�e strengt� of t�e difference between t�e correlation coefficients obtained t�e two age groups was assessed using t�e VassaStats website for statistical omputation. sults Sixty-four 4-year-olds and 68 eig�t-year-olds completed all aspects of t�e study and provided ink samples. T�ree 4-year-olds and five 8-year-olds did not meet t�e inclusion riterion of a urinary flow rate of 5ml/�our and 9 ml/�our respectively and t�ey were excluded from furt�er analysis. Overall, 48% of parents/legal guardians �ad received eit�er no education �15% and 14% for parents of 4- and 8-year-olds respectively) or education to ry sc�ool level only �34% and 33% for parents of 4- and 8-year-olds respectively). T�e n �SD) age, �eig�t and weig�t for 4 year-olds �n=61; 34 male, 27 female) was 4.5 �0.2) ears, 1.0 �0.1) m and 15.5 �2.0) kg respectively; for 8 year-olds �n =64; 33 male, 31 female was 8.6 �0.3) years, 1.2 �0.1) m and 22.3 �3.2) kg. egarding t�e F assay met�ods used, t�e Incurred Sample �e-analysis for t�e 10% samples of water, food, drink and toot�paste s�owed t�at t�e F concentrations of t�e re-analysis were wit�in 20% of t�e averaged concentrations between original and repeat measurements wit� no statistically significant difference in F concentration between test and-test �mean difference was 0.007 mg/l). In addition, t�e mean recovery of F added to t�e samples was wit� a range from 90% to 96%, representing an acceptable

9 level of reliability and good validity
level of reliability and good validity for t�e F analysis met�od used. ��8 Study participants consumed 31 food and drink types but 2 pooled samples �comprising steamed vegetables, and commercially available powdered milks added to tea, c�ocolate tc.) from a low F area �actual drinking water 0.6 mg ppm F; actual cooking water 0.6 F) were not provided by parents/guardians. T�ese 2 missing food and drink samples presented 0.4% of t�e total weig�t of consumed food and drink and contributed 0.03% to t�e estimated daily dietary F intake from diet. T�e F concentration of t�e common community ground water sources ranged from 0.06 - 3 F and t�e 4 study areas were selected and categorised based on t�eF concentrations. Across t�e 4 study areas, t�e mean �SD) F concentrations of actual drinking and cooking waters consumed were 0.76 �0.90) and 0.68 �0.80) ppm F respectively, w�ile t�e median ange) values for bot� waters were 0.40 �0.10 4.00) ppm F. No study participants took any F tablets or supplements; diet and inadvertent toot�paste ingestion were t�eir only sources of F intake. 4-year-olds ble 1 s�ows t�e estimated mean �SD) TDFI, DUFE, DF�, FUFE and FF� of 4 year-olds by area. T�e mean �SD) TDFI ranged from 0.050 �0.019) mg/kg bw/d in t�e urban lower F rea to 0.385 �0.184) mg/kg bw/d in t�e rural �ig�er F area. Diet was t�e predominant source , estimated as contributing an overall mean �SD) of 71% �19) of t�e TDFI �data not s�own) wit� a range from 93% �6) in t�e rural �ig�er F area to 54% ) in t�e rural lower rea. Overall, 100% of t�e 4 year olds used toot�paste wit� a mean �SD) of 1.23�0.42) s per day and 0.52�0.28) g of toot�paste used per brus�ing �data not s�own), wit� no erences between rural and urban communities. Considering urinary F excretion, t�e overall mean �SD) 24� urine volume was 393 �197ta not s�own) and DUFE ranged from 0.021 �0.010) mg/kg bw/d in t񨀆

10 9;e rural lower F area 0.053 �0
9;e rural lower F area 0.053 �0.030/kg bwt�e rural �ig�er F area �Table 1). T�e �ig�est mean DF� �0.293 mg/kg bw/d) andlowest mean FUFE �18%) was estimated for t�e rural �ig�er F area. T�e mean �SD) FF� ranged from 28% in t�e urban �ig�er F area to 72% in t�e rural �ig�er F area. Comparison among t�e 4 study areas indicated no difference in total daily F intake from toot�paste ingestion, w�ereas F intake from diet and TDFI w�ig�er �p<0.001) in t�e rural �ig�er F area compared wit� t�e ot�er 3 areas �Table 1). T�ere was no difference in DUFE ��9 between t�e urban and rural �ig�er F areas, w�ile was lower in t�e two lower F areas t�an t�e rural �ig�er F area �p<0.01). In addition, DF� was �ig�er in t�e rural �ig�er F area r 3 areas �p<0.001). lower FUFE �p=0.029) and �ig�er FF� �p=0.027) w in t�e rural �ig�er F area t�an in t�e urban �ig�er F area, but neit�er FUFE nor FF� ered among t�e ot�er areas. 8-year-olds ta on t�e estimated mean �SD) TDFI, DUFE, DF�, FUFE and FF� of 8-year-olds by area presented in Table 2. T�e mean �SD) TDFI ranged from 0.043 �0.016) mg/kg bw/d in t�e rural lower F area to 0.326 �0.128) mg/kg bw/d in t�e rural �ig�er F area. Diet was t�e dominant source of F, estimated as contributing an overall mean �SD) of 74% �17) of t�e FI wit� a range from 94% �6) in t�e rural �ig�er F area to 67% �16) in t�e urban �ig�er F rea. Overall, 98% of t�e 8 year olds used toot�paste wit� a mean �SD) of 1.08 �0.32) s per day and 0.57�0.27) g of toot�paste used per brus�ing �data not s�own), wit�

11 ; no erences between rural and urban com
; no erences between rural and urban communities. overall mean �SD) 24� urine volume was 618 �337) m �data not s�own). T�e mean DUFE s�owed a wide range from 0.012 �0.007) mg/kg bw/d in t�e rural lower F area to 0.018) mg/kgbw/d in t�e rural �ig�er F area �Table 2). T�e mean �SD) estimated DF� was �ig�est in t�e rural �ig�er F area �0.249 mg/kg bw/d) and lowest in urban lower F area 1 mg/kg bw/d). Consequently, t�e lowest estimated mean FUFE was r t�e rural �ig�er F area; t�e �ig�est being found t�e urban lower F area at 56%. As a result of t�is, �e estimated mean �SD) FF� ranged from 34% in t�e urban lower F area to 73% in t�e rural �er F area. Comparison among t�e 4 study areas s�owed no differences in total daily F intake from toot�paste ingestion, w�ile F intake from diet was �ig�er �p<0.001) in t�e rural �ig�er F area in t�e ot�er 3 areas �Table 2). TDFI, UFE and DF� were �ig�er in t�e rural �ig�er F rea t�an in t�e ot�er 3 areas �p<0.001), w�ile t�ere was no difference in FUFE between n and rural �ig�er F areas and between t�e urban �ig�er and lower F areas. FUFE was �er in t�e urban lower F t�an t�e rural �ig�er F area �p=0.001) and t�e rural lower F area �p=0.017). However, FF� was lower in t�e urban lower F t�an t�e rural �ig�er F �p=0.001) and rural lower F area �p=0.021). Comparison between 4- and 8-year-olds ��10 &#x/MCI; 0 ;&#x/MCI; 0 ;Table 3 presents means and 95% confidence intervals for differences in TDFI, DUFE, DF�, UFE, FF� and urine volume between 4- and 8-year-olds. Total daily F intake from �paste ingestion was �ig�er �p =0.002) in 4-year-olds �0.021 mg/kg bw/d) t�an in 8-year-/kg bw/d). Howe

12 ver, t�ere was no difference in
ver, t�ere was no difference in TDFI between t�e two age groups. Alt�oug� t�ere was no difference in urine volume, adjusted for body weig�t, between 4- and 8-year-olds, t�e DUFE was �ig�er �p=0.013) in 4-year-olds �0.032 mg/kg bw/d) t�an -year-olds �0.022 mg/kg bw/d). T�ere were no differences in FUFE or FF� between t�e ge groups. �elations�ips between t�e key variables Figure 1 illustrates t�e relations�ips between TDFI and DUFE for bot� age groups. T�e slope and intercept of t�e linear relations�ip were 0.046 and 0.026 �respectively) for 4-year-olds nd 0.085 and 0.013 for 8-year-olds. T�e positive correlation between TDFI �mg/kg bw/d) nd DUFE �mg/kg bw/d) was weak for 4-year-olds �r = �0.29), but strong for 8-year-olds �r = �0.64). T�ere was a statistically significant difference 0.012) in t�e correlation coefficient estimated for 4 and 8 year-olds. T�e linear relations�ips between TDFI and DF� for bot� age groups are s�own in Figure 2. very strong positive correlation between TDFI �mg/kg bw/d) and DF� �mg/kg bw/d) was en for bot� age groups: 4-year-olds �r = �0.98), 8-year-olds �r = �0.99). T�ere was no tically significant difference �p=0.06) in t�e correlation coefficient estimated for 4 and 8 year-olds. Alt�oug� t�e intercept of t�e linear relations�ip differed between t�e two age roups �-0.026 in 4-year-olds and -0.013 in 8-year-olds), t�e slopes were very similar �0.854 and 0.815, respectively). Figure 3 presents t�e association between estimated FF� wit� TDFI for bot� age groups. T�e F� increased wit� increasing TDFI and plateaued at TDFI values greater t�an approximately 0.1 mgF/bw/d for bot� age groups. ussion T�is is t�e first report of estimated F intake and excretion in c�ildren in Nigeria. In addition, to t�e best of our knowledge, t�is paper is th

13 0069;e first to report t�e rela
0069;e first to report t�e relations�ip between total intake �from diet and toot�paste ingestion) and urinary retion in 8--olds. A omparison of t�ese F metabolism variables between two age groups living under similar ��11 &#x/MCI; 0 ;&#x/MCI; 0 ;environmental conditions is also unique. W�en adjusted for body weig�t, t�e study found �er urinary F excretion in 4-year-olds compared wit� 8-year-olds, despite no difference in total daily F intake between t�e two age groups, but no difference in F retention. T�ese �anges in t�e pattern of F intake and excretion seen as a c�ild develops from t�e primary xed dentition stages may be relevant w�en seeking to quantify dental fluorosis risk in Alt�oug� t�e range of F concentrations of community ground water supplies sampled was 0.03 to 3.0 ppm F, t�e drinking and cooking waters actually consumed by participants were to 4.0 ppm F; wit�in t�e range of 0.03-6.7 ppm 27 previously reported in Nigeria. T�e F concentrations of actual drinking and cooking waters consumed varied slig�tly from of t�e local community water supplies w�ic� �ad been collected from s�allow wells and aquifers and used to categorise study locations into �ig� and low F water areas. oncentrations of s�allow wells can s�ow �ig� variability, some of w�ic� is due to seasonal differences wit� lower F concentrations found during rainy seasons. T�e actual drinking and cooking waters consumed s�owed less variation in F concentration between areas w�en nalys for fluoride. T�is was primarily because participants¶ individual drinking waters in rticular, were less likely to �ave been obtained from t�e community water supply, but rat�er were commercially purc�ased waters, sold in sac�ets; a common practise in Nigeria to try and optimise drinking water quality. As a result, c�ildren in t�ree of t�e four areas were exposed to fairly similar drinking water F concentrations. Diet and toot�paste ingestion were t�e primary sources of F intake, wit� diet being t�e major component

14 of TDFI for bot� age groups &#x
of TDFI for bot� age groups �Table 3). T�e 71% contribution of diet to TDFI, for Nigerian 4-year-olds corresponds to t�e 70% reported for US c�ildren aged 4 years 28re are no data on TDFI in 8-year-olds, to compare our results wit�, �owever t�e literature wide variation in t�e contribution of diet to TDFI ranging from 88% for 6-year-olds in Iowa 28 to 31% for 4±5-year-olds in Puerto �ico . Differences in c�ildren¶s age, dietary �abits and pattern as well as met�ods of data collection could account for t�e differences in ry contribution to TDFI seen in t�ese studies. n terms of t�e type and constituents of diet seen, solid food components most commonly omprised starc�y staple foods cooked wit� water for prolonged periods,w�ic� made a substantial contribution to F intake, particularly in t�e �ig�er F rural area. A limitation of t�is y was t�at fluoride intake from toot�paste was determined visually by parents¶ ��12 &#x/MCI; 0 ;&#x/MCI; 0 ;questionnaire responses to diagrams depicting t�e amount of toot�paste routinely used. T�e stimated 41% of dispensed toot�paste ingested was based on t�at reported for 4-6 year-olds �e UK 21 and Iranian 4-year-olds 22. However, t�ere was no evidence to suggest t�at t�is s any different in Nigeria w�ere toot�paste use is widespread. T�is approac� to estimation of toot�paste ingestion can be useful for informing epidemiological studies involving large numbers of people w�ere individual toot�brus�ing be�aviours cannot be observed directly and w�ere estimates are required. Nevert�eless, differences in quantities and F contents of �pastes used and in toot�brus�ing �abits can impact comparisons wit� ot�er studies and efore any extrapolation s�ould be cautious. T�e contribution of diet to TDFI �74%) for Nigerian 8-year-olds was slig�tly �ig�er t�an for 4-year-olds, reflecting t�e lower contribution t�roug� toot�paste ingestion for t�e older age roup. C

15 80069;ildren younger t�an 6 yea
80069;ildren younger t�an 6 years are less able to fully control t�eir swallowing reflex resulting in unintentional swallowing of more toot�paste compared wit� older c�ildren T�e present study demonstrated t�e impact of F concentration of water on TDFI and consequently UFE, TDF�, FUFE and FF� �Tables 1-3). In 4-year-olds w�o consumed drinking and cooking waters wit� a median F concentration of ≤ ppm �i.e. urban �ig�er area and urban and rural lower F areas) �Table 1), t�e mean TDFI was wit�in t�e um¶ range of 0.05-0.07 mgF/kg bw/d for maximum benefit in terms of caries reduction wit� minimised risk of dental fluorosis. However, for t�ose four year olds in t�e rural �ig�er area t�e mean TDFI was 0.385 mg/kg bw/d, well above t�e UL of 0.1 mgF/kg bw/d, and y putting t�e c�ildren at greater risk of dental fluorosis. T�e same trend was also rved for 8-year-olds �Table 2); a TDFI range of 0.043-0.057 mg/kg bw/d for t�ose c�ildren receiving drinking and cooking waters wit� median F concentration of ≤ 0.5 ppm versus a TDFI of 0.326 mg/kg bw/d for t�ose receiving drinking and cooking waters wit� n �range) F concentration of 0.3 �3.0) and 0.47 �< 3.0) ppm. In t�e present study, t�e overall mean TDFI of 4-year-olds �0.137 mg/kg bw/d) was only slig�tly �ig�er t�an 8-year-olds �0.106 mg/kg bw/d), w�ereas t�e mean DUFE was tically significantly �ig�er in 4-year-olds �Table 3). T�e estimated intercepts, presented in Figure 1, clearly indicate t�at in t�e absen of any F exposure, t�e DUFE for 4-year-old c�ildren was twice t�at of t�e 8-year-olds �0.026 mg/kg bw/d for 4-year-olds vs 0.013 mg/kg or 8-year-olds) w�ic� could be explained by t�e type/form of ingested F and its ilability. In general, t�e bioavailability of F from toot�paste �eit�er NaF or S�FP) is ��13 &#x/MCI; 0 ;&#x/MCI; 0 ;�ig�er t�an t�at from a mixed die

16 t. T�e amount of F intake from
t. T�e amount of F intake from toot�paste ingestion in 4-year-olds �0.021 mg/kg bw/d) was �ig�er t�an t�at in 8-year-olds �0.014 mg/kg bw/d) �Table t�erefore, t�e amount of absorbed F would �ave been �ig�er in 4-year-olds compared wit� 8-year-olds due to t�e �ig�er bioavailability of t�e F from toot�paste. indicates bioavailability is a �ig�ly relevant factor in body F burden w�en evaluating risk of dental fluorosis, alongside t�e actual amount of fluoride intake. �ysical activity/sedentary be�aviour and/or skeletal development stage impacts DUFE, wit� renal clearance of F declining wit� increasing p�ysical activity 31. A recent study in Sout� rica, reported an increase in activity level wit� age; wit� 9-11 year-old boys and 12-14 ear-old girls more p�ysically and aerobically active t�an boys and girls aged 5-6 years T�e significantly lower DUFE in 8-year-olds in comparison wit� 4-year-olds, observed in t�e present study, could t�erefore relate to t�e effect of �ig�er level of p�ysical activity on renal clearance in t�e older c�ildren. reater rates of F uptake into newly formed bones �i.e. F retention) occurs during periods of pid growt� . Daily skeletal gains of calcium wit� age seen from birt� to puberty follow a ¶ s�ape wit� t�e peak gains being during t�e first mont�s of life and t�en again during t�e dolescent growt� spurt, w�ilst t�e lowest gain is around 3-4 years of age Alt�oug� interpretation s�ould be cautious in view of different populations and met�ods used, w�en t�e FF� data �% F retention) were plotted against age �Figure 4), for 8-year-olds toget�er wit� ilar but limited data in t�e literature for infants , 3-year-olds 34, 4-6 year-olds , 7-year-36 and 11-14 year olds , t�e resultant grap� also s�ows a similar µV¶ s�ape for fluoride. T�e observed overall mean FF� of 46% for 4 ye

17 ar-olds was lower t�an t栀
ar-olds was lower t�an t�e 55% of TDFI stimated to be retained by pre-sc�ool c�ildren but �ig�er t�an t�e 11% and 15% reported 4 year-old Iranian and 3-4 year-old US c�ildren respectively. In contrast, for 8 year-t�e estimated FF� at 54% was similar to t�e 54% reported among 4-5 year-old C�ilean �ildren 23. present study s�owed a positive linear relations�ip between TDFI and DF� �Figure 2) bot� age groups, wit� similar slopes0.854 and 0.815 for 4- and 8-year-olds, respectively, � implies a �ig�er F retention wit� increasing F intake. However, Figure 3 suggests t�at t�e FF� reac�es a limiting constant proportion of 80% above a TDFI of approximately 0.1 mg/kg bw/d, a value similar to t�e suggested Tolerable Upper Intake Level �UL) for F onstant proportion of 80% for FF� seen in bot� age groups above t�is TDFI t�res�old is ��14 &#x/MCI; 0 ;&#x/MCI; 0 ;�ig�er t�an t�e 50% suggested by Ekstrand for infants aged 2-5 mont�s and t�e 55% found by Villa et al in a broader age group of c�ildren, and s�ould be considered furt�er w�en ssessing any increased risk of fluorosis from excessive F accumulation in �ard tissues due to t�e �ig�er levels of F exposure seen in some of t�ese Nigerian c�ildren. In t�e present study it was useful to be able to compare t�e FUFE between t�e two age roups subjected to similar environmental influences including diet. T�e lower FUFE seen in t�e older c�ildren is consistent wit� results from ot�er studies of differing age groups in t�e me environment, in w�ic� older c�ildren retain a greater % of t�eir F intake up to t�res�old levels . T�e present study eliminated some of t�e differences in confounding variables seen fferent populations, providing a clearer picture of t�e difference in FUFE �and t�erefore y F retention) seen age. In 4-year-olds, t�ere was no significant diffe

18 rence in FUFE among t�e tڀ
rence in FUFE among t�e t�ree areas in w�ic� t�e median �range) of drinking and cooking waters was 0.07 �≤ 0.6) ppm F but FUFE in t�ese reas was significantly �ig�er t�an in t�e rural �ig�er F area wit� 0.3 �≤ 4.0) ppm F waters. ever, interestingly, in 8-year-olds, despite no difference in TDFI and DUFE among t�e e areas wit� median �range) water F of 0.08 �≤ 0.5) ppm F, t�e FUFE was significantly �er in t�e urban lower F- �median �range); < 0.5 �1.0) ppm Ft�an in t�e rural lower F- �median �range); < �≤ 0.5) ppm and rural �ig�er F �median �range) 0.47 �≤ 3.0) ppm F) areas. se findings may be explained by possible differences in dietary omposition and patterns between rural and urban areas as well as between 4- and 8-year-, all of w�ic� affect F absorption and excretion. For example, an increase in t�e proportion of TDFI excreted in t�e urine �as been s�own for groups consuming rice-based ompared wit� sorg�um-based diets 39. It �as also been suggested t�at �ig� dietary concentrations of certain cations suc� as calcium in milk can reduce t�e extent of F absorption . Furt�ermore, F absorption may increase w�en diets ric� in protein and fats are onsumed since t�ey reduce t�e rate of gastric emptying, w�ile diets w�ic� are primarily vegetarian alkalinise t�e urine resulting in decreased reabsorption of F 41. In t�e present y, no comparisons on dietary composition and �abits were made between rural and urban areas between 4- and 8-year-olds. T�e effect of dietary composition on F absorption and xcretion merits furt�er investigation, not only in Nigeria, but also globally. T�e overall mean FUFE of t�e Nigerian 4-year-olds was 44%; very close to t�e 42% reported for 4-year-olds . To t�e best of our knowledge, t�ere are no FUFE data in t�e literature ��15 &#x/MCI; 0 ;&#x/MCI; 0 ;for 8-year-olds. However, t�e overall mean FUFE of t

19 �e Nigerian 8-year-olds, ʀ
�e Nigerian 8-year-olds, �36%) was ilar to t�e FUFE of 35% reported for C�ilean 11-14-year-olds . present study found a positive linear relations�ip between TDFI and DUFE �Figure 1) for bot� age groups, �owever, t�e slopes of t�e correlation differed �0.046 and 0.085 for 4- and 8-year-olds, respectively). T�is finding provides furt�er evidence t�at DUFE can be used to stimate TDFI; similar t�e study by Villa and co-workers relations�ip between UFE and TDFI was examined using previously publis�ed data on F intake and xcretion in c�ildren younger t�an 7 years old, and s�owed t�at, on average, 35% of TDFI s excreted in t�e urine.T�e findings of t�e present study supplement t�is latter study, ng gaps in t�e dataset and suggest t�at: �F intake and excretion data of 8-year-olds fits well wit�in t�e distribution of corresponding data for younger age groups; and �b) similarities exist in F intake and excretion data, collected from developing and developed countries,y due to globalisation of t�e food system and disappearance of food traditions in developing countries in recent years. Acknowledgements T�e aut�ors t�ank t�e c�ildren and t�eir parents/guardians for t�eir participation and cooperation. Support from t�e Commonwealt� Sc�olars�ip Commission and Centre for Oral Healt� �esearc�, Newcastle University is gratefully acknowledged. ��16 en-US&#x/Lan;&#xg 00;&#x/Lan;&#xg 00;�eferences 1. Bronckers AL, Lyaruu D�, Denbesten PK. Critical review in oral biology and medicine: T�e impact of fluoride on ameloblasts and t�e mec�anisms of enamel fluorosis. J Dent �es 2009;88:877-93.2. �obinson C, Connell S, Kirk�am J, Brookes SJ, S�ore �C, Smit� A�. T�e effect of fluoride on t�e developing toot�. aries �es 2004;38:268-76. Institute of �edicine. Dietary �eference Intakes for Calcium, �agnesium, Vitamin D, and Fluoride. Was�ington DC:

20 National Academy Press; 1997. guire A,
National Academy Press; 1997. guire A, =o�ouri FV, Hindmarc� PN, Hatts J, �oyni�an PJ. Fluoride intake and urinary excretion in 6- to 7-year-old c�ildren living in optimally, sub-optimally and non-idated areas. Community Dent Oral Epidemiol 2007;35:479-88. 5. �aguire A, =o�oori FV. Floride balance in infants and young c�ildren in t�e UK and its linical relevance for t�e dental team. Dent J 2013;214:587-936. W�itford G�. T�e metabolism and toxicity of fluoride. �onograp�s in oral science Basel: rger; 1996. 7. World Healt� Organization. Basic �et�od of Assessment of �enal Fluoride Excretion in Community Prevention Programmes for Oral Healt�, Geneva, World Healt� Oraganization; 2014. a A, Anabalon �, =o�ouri V, �aguire A, Franco A�, �ugg-Gunn A. �elations�ips tween fluoride intake, urinary fluoride excretion and fluoride retention in c�ildren and adults: an analysis of available data. Caries �es 2010;44:60-8. ng'a P�, Valder�aug J. Prevalence and severity of dental fluorosis in primary sc�oolc�ildren in Nairobi, Kenya. Community Dent Oral Epidemiol 1993;21:15-18. ongdem JG, Aderinokun GA, Ubom GA, Srid�ar �K, Selkur S. Dental fluorosis and fluoride mapping in Langtang town, Nigeria. Afr J �ed �ed Sci 2001;30:31-4. irempong C, Nsia� K, Awunyo-Vitor D, Dongsogo J. Soluble fluoride levels in drinking ter-a major risk factor of dental fluorosis among c�ildren in Bongo community of G�ana. G�ana �ed J 2013;47:16-23. arvis HG, Heslop P, Kisima J, Gray WK, Ndossi G, �aguire A et al. Prevalence and etiology of juvenile skeletal fluorosis in t�e sout�-west of t�e Hai district, Tanzania - community-based prevalence and case-control study. Trop �ed Int Healt�. 2013;18:222-13. S�orter JP, �assawe J, Parry N, Walker �W. Comparison of two village primary sc�ools in nort�ern Tanzania affected by fluorosis. Int Healt� 2010;2:269-74. ��17 en-US

21 &#x/Lan;&#xg 00;&#x/Lan;&#xg 00;14. Worl
&#x/Lan;&#xg 00;&#x/Lan;&#xg 00;14. World �edical Association Declaration of Helsinki Et�ical Principles for �edical �esearc� involving �uman subjects. J Aed Assoc2191-4.15. Ibiyemi O, =o�oori FV, Valentine �, Kometa S, �aguire A. Prevalence and extent of enamel defects in t�e permanent teet� of 8-year-old Nigerian c�ildren. Community Dent Oral Epidemiol-6216. National Population Commission and ICF �acro. Nigeria Demograp�ic and Healt� urvey 2008: Key Findings. 2009. 17. �ankin SJ, Levy S�, Warren JJ, Gilmore JE, Broffitt B. �elative validity of an FFQ for ssessing dietary fluoride intakes of infants and young c�ildren living in Iowa. alt� Nutr 2011;14:1229-36. 18. Levy S�, Warren JJ, Davis CS, Kirc�ner HL, Kanellis �J, Wefel JS. Patterns of fluoride ake from birt� to 36 mont�s. J Pub Healt� Dent2001;61:70-7. 19. �artínez-�ier EA, Cury JA, Heilman J�, Katz BP, Levy S�, Li � et al. Development of old standard ion-selective electrode-based met�ods for fluoride analysis. Caries �es2011;45: 3-12. 20. =o�oori FV, �aguire AA. Development of a Database of t�e Fluoride Content of elected Drinks and Foods in t�e UK. Caries �es 2016;50: 331-6. 21. =o�oori FV, Duckwort� ��, Omid N, O'Hare WT, �aguire A. Fluoridated toot�paste: ge and ingestion of fluoride by 4- to 6-old c�ildren in England. Eur J Oral Sci415-21.22. =o�ouri FV, �ugg-Gunn AJ. Total fluoride intake and urinary excretion in 4-year-old ranian c�ildren residing in low-fluoride areas. Br J Nutr 2000;83:15-25. 23. Villa A, Anabalón �, Cabezas L. T�e fractional urinary fluoride excretion in young �ildren under stable fluoride intake conditions. Community Dent Oral Epidemiol2000;28:344-55. 24. Ekstrand J, Hardell LI, Spak CJ. Fluoride balance studies on infants in a 1-ppm-water-ide area. Caries �es 1984;18:87-92. 25. =ar JH. Biostatistical Analysis. New Jersey: Prentice Hall; 1999. 26. Significance of t�

22 ;e difference between two correlation Co
;e difference between two correlation Coefficients. p://vassarstats.net/rdiff.�tml . Accessed 27 July 2017. Akpata E, Danfillo I, Oto� E, �afeni J. Geograp�ical mapping of fluoride levels in water sources in Nigeria. Afr Healt� Sci 2009;9:227-33. 28. Levy S�, Warren JJ, Broffitt B. Patterns of fluoride intake from 36 to 72 mont�s of age. Healt� Dent2003;63:211-20. ��18 en-US&#x/Lan;&#xg 00;&#x/Lan;&#xg 00;29. �ojas-Sanc�ez F, Kelly SA, Drake K�, Ec�ert GJ, Stookey GK, Dunipace AJ. Fluoride ake from foods, beverages and dentifrice by young c�ildren in communities wit� negligibly and optimally fluoridated water: a pilot study. Community Dent Oral 1999;27:288-97. 30. �aternal and C�ild Healt� Bureau ��CHB) Expert Panel. Topical Fluoride ecommendations for Hig�-�isk C�ildren Development of Decision Support �atrix. �ecommendations from �CHB Expert Panel. Was�ington, DC, Altarum Institute; 2007. 31. =o�oori FV, Innerd A, Azevedo LB, W�itford G�, �aguire A. Effect of exercise on ide metabolism in adult �umans: a pilot study. Scientific �eports 2015;5:16905. 32. �innaar E, Grant CC, Fletc�er L. P�ysical activity of c�ildren from a small rural town, out� Africa. Sout� African Family Practice 2016;58:68-73. 332. Kanis JA, Passmore �. Calcium supplementation of t�e diet -�ed J 1989;298:137-34. =o�ouri FV, Swinbank C�, �aguire A, �oyni�an PJ. Is t�e fluoride/creatinine ratio of a urine sample indicative of 24-� urinary fluoride? ommunity Dent Oral Epidemiol2006;34:130-8. 35. Omid N, �aguire A, O'Hare WT, =o�oori FV. Total daily fluoride intake and fractional ry fluoride excretion in 4- to 6-year-old c�ildren living in a fluoridated area: weekly riation? Community Dent Oral Epidemiol 2016;45:12-9. 36. =o�oori F, Walls �, Teasdale L, Landes D, Steen IN, �oyni�an P et al. Fractional urinary fluoride excretion of 6-7-year-old cڀ

23 069;ildren attending sc�ools in
069;ildren attending sc�ools in low-fluoride and turally fluoridated areas in t�e UK. Br J Nutr 2013;109:1903-9.37. Villa A, Cabezas L, Anabalon �, Garza E. T�e fractional urinary fluoride excretion of dolescents and adults under customary fluoride intake conditions, in a community wit� 0.6-mg F/L in its drinking water. Community Dent alt� 2004;21:11-8. 38. Brunetti A, Newbrun E. Fluoride balance of c�ildren 3 and 4 years old. aries �es1983;17:171. 39. B�argavi V, K�andare AL, Venkaia� K, Sarojini G. �ineral content of water and food in otic villages and prevalence of dental fluorosis. Biol Trace Element �es2004;100:195-203. . W�itford G�.Effects of plasma fluoride and dietary calcium concentrations on GI absorption and secretion of fluoride in t�e rat. Calcif Tissue Int1994;54:421-25. 41. Buzalaf �A� and W�itford G�. Fluoride metabolism. Switzerland: Karger; 2011. 19 ��20 en-US&#x/Lan;&#xg 00;&#x/Lan;&#xg 00;Legends for Tables and Figures. Table 1. Estimated �ean �SD) fluoride intake �mg/kg bw/d), urinary fluoride excretion /kg bw/d), fluoride retention �mg/kg bw/d), fractional urinary fluoride excretion �%) and actional fluoride retention �%) among 4-year-olds �n=61) by area. Table 2. Estimated �ean �SD) fluoride intake �mg/kg bw/d), urinary fluoride excretion /kg bw/d), fluoride retention �mg/kg bw/d), fractional urinary fluoride excretion �%) and actional fluoride retention �%) among 8-year-olds �n=63) by area. Table 3. �ean and 95% confidence interval of difference in fluoride intake �mg/kg bw/d), ry fluoride excretion �mg/kg bw/d), fluoride retention �mg/kg bw/d), fractional urinary ide excretion �%) and fractional fluoride retention �%) between 4- �n=61) and 8-year-63). Figure 1: �elations�ip between total daily F intake �TDFI: mg/kg bw/d) and daily urinary F excretion �DUFE: mg/kg bw/d) for 4-year-olds �n=61) and 8-year olds �n=63). igure 2: �elations�ip between total daily F intake �TDFI: mg/kg bw/d) and daily F retention &#x

24 280029;DF�: mg/kg bw/d) for
280029;DF�: mg/kg bw/d) for 4-year-olds �n=61) and 8-year olds �n=63). Figure 3: �elations�ip between total daily F intake �TDFI: mg/kg bw/d) and fractional F retention �FF�: %) for 4-year-olds �n=61) and 8-year olds �n=63). Figure 4. Fluoride retention �%) by age, based on data from t�e present study and relevant literature. 24, 34-37 ��21 &#x/MCI; 0 ;&#x/MCI; 0 ;Table 1.Estimated �ean �SD) fluoride intake �mg/kg bw/d), urinary fluoride excretion �mg/kg bw/d), fluoride retention �mg/kg bw/d), ractional urinary fluoride excretion �%) and fractional fluoride retention �%) among 4-year-old�n=61) by area. Area Hig�er F area Lower F area Urban 1 �n=16) �ural �n=15) Urban �n=17) �ural �n=13) Total daily F intake �mg/kg 0.059 �0.029) 0.385 �0.184) 0.050 �0.019) 0.062 �0.022) Total daily F intake from diet �mg/kg 0.041 �0.026) 0.362 �0.181) 0.033 �0.012) 0.033 �0.012) Drinks 0.020 �0.025) 0.094 �0.097) 0.003 �0.004) 0.010 �0.011) 0.021 �0.007) 0.268 �0.158) 0.028 �0.012) 0.023 �0.008) Contribution of diet to total daily F intake �%) 67 �17) 93 �6) 68 �15) 54 �16) Total daily F intake from toot�paste ingestion �mg/kg 0.018 �0.011) 0.022 �0.014) 0.016 �0.011) 0.029 �0.016) Daily u rinary F excretion �mg/kg 0.031 �0.028) 0.053 �0.030) 0.025 �0.025) 0.021 �0.010) Daily F �etention �mg/kg 0.022 �0.037) 0.293 �0.178) 0.020 �0.028) 0.035 �0.021) Fractional Urinary F excretion �%) 62 �52) 18 �13) 54 �58) 37 �20) Fractional F �etention �%) 28 �52) 36 �58) 53 �20) Variables wit� different letters are statistically significantly different at p<0.05 �post-�oc test). *ANOVA ± No significant difference. �See Supplementary material available). Community ground water �ppmF) = 0.8-1.0; �ange ⠀

25 29;median) drinking and cooking water &#
29;median) drinking and cooking water �ppmF), respectively: ≤0.43 �0.01) ppmF, ≤0.40 �0.03) ppmF Community ground water �ppmF) = 2.0-3.0; �ange �median) drinking and cooking water �ppmF), respectively: ≤3.67 �0.3) ppmF, ≤4.00 �0.33) ppmF ommunity ground water �ppmF) = 0.04-0.07; �ange �median) drinking and cooking water �ppmF), respectively: ≤0.08 �<0.01) ppmF, ≤2.00 �<0.01) ppmF ommunity ground water �ppmF) = 0.06-0.07; �ange �median) drinking and cooking water �ppmF), respectively: ≤0.60 �0.07) ppmF, ≤0.40 �0.01) ppmF ��22 &#x/MCI; 0 ;&#x/MCI; 0 ;Table 2. Estimated �ean �SD) fluoride intake �mg/kg bw/d), urinary fluoride excretion �mg/kg bw/d), fluoride retention �mg/kg bw/d), ractional urinary fluoride excretion �%) and fractional fluoride retention �%) among 8-year-olds �n=63) by area. Area Hig�er F area Lower F area Urban �n=13) �ural �n=13) Urban �n=17) �ural �n=20 Total daily F intake �mg/kg 0.057 �0.045) 0.326 �0.128) 0.048 �0.038) Total daily F intake from diet �mg/kg 0.041 �0.039) 0.307 �0.120) 0.033 �0.012) �0.014 Drinks 0.016 �0.024) 0.073 �0.081) 0.003 �0.004) �0.013 0.025 �0.019) 0.234 �0.096) 0.030 �0.039) �0.010 Contribution of diet to total daily F intake �%) 67 �16) 94 �6) 72 �16) 68 �15) Total daily F intake from toot�paste ingestion �mg/kg 0.015 �0.010) 0.019 �0.016) 0.012 �0.007) Daily u rinary F excretion �mg/kg 0.017 �0.014) 0.044 �0.018) 0.022 �0.013) �0.007 Daily F �etention �mg/kg 0.034 �0.041) 0.249 �0.112) 0.021 �0.034) �0.015 Fractional Urinary F excretion �%) 39 �33) 17 �14) 56 �34) 29 �19 Fractional F �etention �%) 51 �33) 73 �14) 34 �34) 61 �19 Variables wit� different letters were statistically significantly

26 different at p<0.05 �post-
different at p<0.05 �post-�oc test). *ANOVA ± No significant difference. �See Supplementary material available).Community ground water �ppmF) = 0.8-1.0; �ange �median) drinking and cooking water �ppmF), respectively: ≤0.2 �0.01) ppmF, ≤0.20 �0.01) ppmF Community ground water �ppmF) = 2.0-3.0; �ange �median) drinking and cooking water �ppmF), respectively: ≤3.0 �0.3) ppmF, ≤3.00 �0.47) ppmF Community ground water �ppmF) = 0.04-0.07; �ange �median) drinking and cooking water �ppmF), respectively: ≤0.2 �0.01) ppmF, ≤1.00 �0.05) ppmF Community ground water �ppmF) = 0.06-0.07; �ange �median) drinking and cooking water �ppmF), respectively: ≤0.50 �0.08) ppmF, ≤0.50 �0.07) ppmF ��23 &#x/MCI; 0 ;&#x/MCI; 0 ;Table 3.�ean and 95% confidence interval of difference in fluoride intake �mg/kg bw/d), urinary fluoride excretion �mg/kgbw/d), luoride retention �mg/kg bw/d), fractional urinary fluoride excretion �%) and fractional fluoride retention �%) between 4- �n=61) and 8-year-olds �n=63). �ean �SD) �ean �95% Confidence interval) of difference P value year �n=61) year �n=63) Total daily F intake �mg/kg 0.137 �0.169) 0.106 �0.130) �0.031 � 0.022, �0.085) Total daily F intake from diet �mg/kg 0.116 �0.167) 0.091 �0.126) �0.025 � 0.028, �0.078) Drinks 0.031 �0.061) 0.024 �0.046) �0.007 � 0.012, �0.026) 0.085 �0.131) 0.67 �0.098) �0.017 � 0.024, �0.059) Contribution of diet to total daily F intake �%) 71 �19) 74 �17) 10, �3) Total daily F intake from toot�paste ingestion �mg/kg 0.021 �0.013) 0.014 �0.010) �0.007 ��0.002, �0.011) Daily urinary F excretion �mg/kg 0.032 �0.027) 0.022 �0.017) �0.010 �+

27 002C002D002E;0.002, �0.
002C002D002E;0.002, �0.018) Daily F �etention �mg/kg 0.091 �0.147) 0.073 �0.107) �0.018 � 0.027, �0.064) Fractional Urinary F excretion �%) 44 �44) 36 �30) �8 � 5, �21) Fractional F �etention �%) 46 �44) 54 �30) 21, �5) Urine volume �ml/kg bw/d) 24.8 �10.0) 27.9 �14.8) 7.6, �1.4) ��24 &#x/MCI; 0 ;&#x/MCI; 0 ;Figure 1: �elations�ip between total daily F intake �TDFI: mg/ bw/d) and daily rinary F excretion �DUFE: mg/ bw/d) for 4-year-olds �n=61) and 8-year olds �n=63). 4-year-olds: DUFE �mg/kg bw/d)=>0.046 x TDFI�mg/kg bw/d)]�0.026; �r) = �0.29, P=0.025. 8-year-olds: DUFE �mg/kg bw/d)=>0.085 FI�mg/kg bw/d)]�0.013; �r) = �0.64, P <0.001. DUFE �mg/kg bw/d) TDFI �mg/kg bw/d) 4-year-olds 8-year-olds 4-year-olds 8-year-olds ��25 &#x/MCI; 0 ;&#x/MCI; 0 ;Figure 2: �elations�ip between total daily F intake �TDFI: mg/kg bw/d) and daily F retention �DF�: mg/kg bw/d) for 4-year-olds �n=61) and 8-year olds �n=63). 4-year-olds: DF� �mg/kg bw/d)=>0.854 x TDFI�mg/kg bw/d)] -0.026; �r) = �0.98, P <0.001. 8-year-olds: DF� �mg/kg bw/d)=>0.815 x TDFI�mg/kg bw/d)] -0.013; �r) = �0.99, P <0.001. -0.20 DF� �mg/kg bw/d) TDFI �mg/kg bw/d) 4-year-olds 8-year-olds 4-year-olds 8-year-olds ��26 &#x/MCI; 0 ;&#x/MCI; 0 ;Figure 3: �elations�ip between total daily F intake �TDFI: mg/kg bw/d) and fractional F retention �FF�: %) for 4-year-olds �n=61) and 8-year olds �n=63). -120 -100 -80 -60 -40 -20 FF� �%) TDFI �mg/kg bw/d) 4-year-olds 8-year-olds ��27 &#x/MCI; 0 ;&#x/MCI; 0 ;Figure 4. Fluoride retention �%) by age, inferred from data from t�e present study and elevant literature. 3, 33- 45 50 55 60 0 2 4 6 8 10 12 F Reten