Community Dent Oral Epidemiol 2012; 40: 2636 All rights reserved

2011 John Wiley & Sons A/S

Fluoride content of Ready-toFeed (RTF) infant food and drinks in the UK

Maguire A, Omid N, Abuhaloob L, Moynihan PJ, Zohoori FV. Fluoride content of Ready-to-Feed (RTF) infant food and drinks in the UK. Community Dent Oral Epidemiol 2012; 40: 2636. 2011 John Wiley & Sons A S Abstract Background: The level of Fluoride exposure needed to cause dental uorosis is not known precisely. An awareness of total F intake from all sources, especially during the critical stages of dental development during infancy and early childhood, is important in preventing the development of dental uorosis. Objectives: The aim of the study was to measure F content of ready-to-feed (RTF) infant drinks and foods in the UK. Methods: In total, 122 infant foods were analysed for F concentrations, in triplicate, indirectly by an acid diffusion method and 25 infant drinks analysed directly using an F-ion-selective electrode after addition of TISABIII. Results: The median (range) F concentration was 0.110 (0.0300.221) lg g for breakfast cereals, 0.112 (0.0401.200) lg g for savoury meals, 0.056 (0.0300.379) lg g for desserts, 0.044 (0.0200.191) lg g for fruits, 0.196 (0.0400.397) lg g for baked goods, 0.069 (0.0500.148) lg ml for juices, 0.016 (0.0090.030) lg ml for milks and 0.041 (0.0220.069) lg ml for waters. The median (range) F concentration of all RTF infant foods and drinks by recommended age of consumption was 0.029 (0.0100.245), 0.088 (0.020 0.500), 0.108 (0.1000.510) and 0.108 (0.0601.200) lg g for infants from birth, 4+ month, 6+ month and 10+ month, respectively. Conclusion: The results suggest that the F concentrations of UK-marketed RTF infant foods, drinks and formula milk are not sufciently high to be a risk factor for dental uorosis, if consumption is within the limits recommended for infants and young children.

Anne Maguire1, Narges Omid2, Lamis Abuhaloob1,3, Paula J. Moynihan1,3,4 and Fatemeh V. Zohoori2
1 Centre for Oral Health Research, School of Dental Sciences, Newcastle University, Newcastle, UK, 2School of Health & Social Care, Teesside University, Newcastle, UK, 3 Institute for Ageing and Health, Newcastle University, Newcastle, UK, 4Human Nutrition and Research Centre, Newcastle University, Newcastle, UK

Key words: diet; uoride; infants; readyto-feed Anne Maguire, Centre for Oral Health Research, School of Dental Sciences, Newcastle University, Newcastle upon Tyne, NE2 4BW, UK Tel.: +44 (0) 191 2228564 Fax: +44 (0) 191 2225928 e-mail: A.Maguire@ncl.ac.uk Submitted 10 January 2011; accepted 1 July 2011

Background
Infancy and early childhood is a critical period for tooth development, and exposure to F during this period can be very inuential in terms of oral health. The role of F in preventing dental caries has been well established (1), although its inuence on bone metabolism is less well dened. Following absorption from the gastrointestinal tract, F ingested from the diet or inadvertently ingested during toothbrushing is rapidly incorporated into calcied tissues that contain 99% of body F (1). While F ingestion during tooth development is related to reduced caries experience (2), excessive ingestion during the rst three years of life appears to be most critical in dental uorosis aetiology.

Based on current evidence, the prevalence of dental uorosis has increased in industrialised countries, in both uoridated and nonuoridated areas over the last 2 decades (35). The increased prevalence of dental uorosis appears to be due primarily to inadvertent ingestion of uoridated toothpastes, with inappropriate use of F supplements in areas with a uoridated water supply and an increase in the halo effect, increased consumption of food and drink processed with uoridated water in nonuoridated areas (6). The level of F exposure needed to cause dental uorosis is not known precisely, and there is a relatively narrow gap between an excessive and benecial intake of F. The current consensus is that a total daily F intake of 0.050.07 mg F kg
doi: 10.1111/j.1600-0528.2011.00632.x

26

Fluoride content of infant food and drinks

body weight in children is generally considered optimal for dental health benets and that the total daily F intake should not exceed 0.10 mg F kg body weight to avoid an undesirable degree of dental uorosis (7). This is in line with the UKs Department of Health guidance (8, 9) which states that intakes of F of 0.12 mg kg body weight per day in infants and children between 6 months and 6 years of age are safe and intakes of F of 0.05 mg kg body weight daily in children aged over 6 years and in adults are therefore safe. More recently, The European Food Safety Authority (EFSA) has set the tolerable upper intake level of F as 0.1 mg F kg body weight day in children 18 years (10), while the U.S. Environmental Protection Agency quotes a no observed adverse effect level (NOAEL) for F of 0.06 mg kg bw per day (11, 12). These differences in guidelines for F exposure based on body weight seen internationally highlight the current regulatory differences that exist between countries and regions and reect the complex relationship between dietary and other F sources and skeletal and dental health, especially in young children. To further the understanding of this relationship, the quantication and monitoring of F exposure (13, 14) is essential. The main F sources for young children are inadvertent ingestion of toothpaste, but non-food F supplements, when used, as well as diet (food and drinks) also contribute to total F intake. In contrast, for infants who are not usually exposed to toothpaste use until the rst teeth erupt into the mouth from approximately 6 months of age, the main source of F is diet (15, 16). Infant formula milks and commercially available beverages and foods are the main sources of F before a child has any teeth to brush, and the contribution of dietary F to total ingested F can range from almost 100% in 6-week-old infants to 70% in 3-year-old children (16). However, as a proportion, the dietary contribution is higher in populations not exposed to F in toothpaste (17). In uoridated communities, water has been considered as the primary source of F, not just from drinking tap water, but also because foods consumed are prepared with uoridated water. However, with a trend towards consumption of foods and drinks made outside the home and the increase in the ready-to-eat as well as ready-to-feed (RTF) markets, the impact of the F content of the home water supply on total F intake has lessened and may no longer provide an adequately specic marker of dietary F exposure

(6). Despite falling birth rates in the UK, the infant food market and, within that market, RTF products have shown a substantial increase of 28.2%, in monetary value terms, between 1995 and 2000 (18), with the advantage of their convenience often outweighing the disadvantage of cost for busy families, especially in more afuent developed countries. With globalisation and ongoing conglomeration of food companies, many food products can be found in multiple countries under the same label although their place of manufacture may vary. With the increasing consumption of RTF products and the range in source of manufacture, there is a need to understand their contribution to total F exposure. Food composition tables are produced nationally in the UK, as well as regionally across the world, to provide information on the nutrient value and most trace element contents of foods and drinks commonly consumed as constituents of the diet in these areas (19). However, there are no data on the F contents of foods and drinks in any food composition tables to facilitate estimation of dietary F exposure in any population, although in the US, a national F database that contains F content of selected chief contributors to F intake for the US population has been developed (20). In the UK, there are no data on the F content of infant drinks and foods currently available commercially, although a F database that contains current F values for selected food and drink items that are major F contributors for children is in development (21, 22). The rst infant food F analyses were carried out over 30 years ago, and with regard to RTF infant foods, the most recent information in the UK is now 17 years old. Walters et al. included 20 samples of infant foods in their F analysis of a selected group of British foods as part of a larger study of dietary intake of F in 11 UK cities (23). They reported a mean F concentration of <0.3 lg g for canned meat products, <0.5 lg g for jar meat products, and <0.2 lg g for canned and jar desserts. In 1992, Vlachou et al. (24) reported the F concentration for 83 baby foods and 30 baby drinks purchased in Leeds, England. Of the RTF products, dairy products ranged from 0.02 to 0.18 lgF g, cereals from 0.04 to 0.70 lgF g, vegetable products from 0.03 to 0.48 lgF g, fruits from 0.03 to 0.07 lgF g and desserts from 0.02 to 0.28 lgF g (24). In the same study, an F analysis of 24 RTF infant drinks showed a range of F concentrations from 0.01 to 0.26 lg ml, while the

27

Maguire et al.

F concentration of RTF milks ranged from 0.01 to 0.11 lg ml. In view of the lack of current information regarding the F content of infant foods and drinks available commercially, the trend towards increased consumption of these products and the fact that brands and compositions change over time, this study was designed to undertake a F analysis of a broad selection of UK-manufactured RTF products, classied according to food and drink subgroups, manufacturers and recommended age of consumer.

Materials and methods

Three different batch numbered products of different brands of a convenience sample of 122 infant RTF food and 25 infant RTF drink products were purchased from 14 major supermarkets, grocery stores and health food shops in three cities in the north-east of England between Nov 2008 and May 2009. As such, these products represented the majority of the products available on the UK market at this time. Equal weights volumes of each of the three batches were mixed together, and then four aliquots of approximately 5 g of the mixed samples were taken and stored in F-free zip-locked bags (solid foods) or 7-ml test tubes (liquid samples) and frozen at )20C. Each mixed sample was analysed in triplicate for F concentration, using an indirect hexame-

thyldisiloxane (HMDS)-facilitated acid diffusion method (2527). The 25 infant drinks were analysed, in triplicate, using a direct method and F-ionselective electrode (Orion Research, model 96-09) after addition of TISABIII (27, 28). Reproducibility of the method was assessed by re-analysis of 10% of samples and to check the validity of the analytical method, a known concentration F standard was added to 10% of the samples, prior to re-analysis, in triplicate, to measure F recovery. The analysed items were divided into two major groups: foods and drinks. Within each group, all the food drink items were categorised further into several groups as described in Table 1. The mean F concentration of each infant food drinks sample was entered into a Microsoft Excel spreadsheet. Descriptive analysis was undertaken using an spss statistical package to report mean, median and range of F concentrations of all samples, classied according to food and drink subgroups, their manufacturers and recommended age of consumer.

Results
The mean (SD) recovery of added F to the food drink samples was 98.5% (2.4%), representing good validity for the method of F analysis used. The correlation between analysis and re-analysis of samples was strong (R = 97%) indicating excellent reproducibility.

Table 1. Mean, median and range of F concentrations (lg g or lg ml) for ready-to-feed infant food and drink groups Number of items 6 79 23 6 14 14 6 16 16 9 7 11 10 6 8 11 F Concentration (lg g or lg ml) Range 0.0300.221 0.0401.200 0.0700.271 0.1050.229 0.0431.200 0.0670.450 0.0700.120 0.0400.309 0.0300.379 0.0330.245 0.0300.379 0.0200.191 0.0400.397 0.0220.069 0.0500.148 0.0090.030 Median 0.110 0.112 0.112 0.122 0.147 0.104 0.099 0.130 0.056 0.055 0.089 0.044 0.196 0.041 0.069 0.016 Mean 0.112 0.152 0.125 0.145 0.259 0.141 0.099 0.130 0.103 0.091 0.118 0.061 0.200 0.044 0.076 0.018

Food drink group Food groups Breakfast cereals Savoury meals Chicken & Turkey Fish Meat (lamb, beef, pork) Pasta & spaghetti Cheese & vegetables Vegetables Desserts Dairy Nondairy Fruits Baked goods Drink groups Waters Juices Milks

28

Fluoride content of infant food and drinks

1% 1%

4%

25%

Table 2. Mean, median and range of F concentrations (lg g or lg ml) of ready-to-feed foods and drinks by manufacturer-specied age group. (Each product was categorised once, according to instructions for appropriate age of use given by manufacturer) Age range specied on product F concentration lg ml) Mean 0.059 0.105 0.140 0.184 0.123 Median 0.029 0.088 0.108 0.108 0.100 (lg g Range 0.0100.245 0.0200.500 0.1000.510 0.0601.200 0.0101.200 or

Number of items 14 59 58 16 147

42%

From Birth 4-month+ 6-month + 10-month+ Total

27%

<0.05 (g/g) 0.310.50 (g/g)

0.050.10 (g/g) 0.511.00 (g/g)

0.110.30 (g/g) >1.00 (g/g)

Fig. 1. The proportional distribution of F concentration (lg g) for 147 ready-to-feed infant foods and drinks.

showed the mean F concentration of RTF foods and drinks increased as the suggested age of consumption increased from birth to 10 months onwards (Table 2). However, the range of F concentrations of RTF foods and drinks showed some overlap between age groups.

Concentration of F in RTF food and drink groups

Table 1 presents the median and range of F concentrations for RTF food and drink groups. The proportional distribution of F concentration for the 147 RTF infant foods and drinks is summarised in Fig. 1. Almost half (52%) of the samples were low in F (0.1 lgF g), while 42% ranged from 0.11 to 0.30 lgF g and only 1% contained greater than 1.0 lgF g. There was a wide range in the concentrations of F found between and within the different food (range; 0.0201.200 lg g) and drink (range; 0.009 0.148 lg g) groups, with savoury meals showing the largest variation. Within the savoury meals food group, the meat subgroup contained the highest concentration of F (median; 0.147 lg g). This was followed by the baked goods food group with a median F concentration of 0.196 lg g. With regard to the drinks as a whole, the F concentration (median 0.028 lg ml) was lower than that of the food groups (median; 0.109 lg g) with the milks group containing the lowest concentrations of F, while the juices contained the highest F concentrations (Table 1).

Manufacturer
No manufacturer labelled the F concentration of its product, and the place of manufacture was rarely labelled, with place of distribution (UK) being more likely to be recorded on the product. When the F concentration of RTF milks was considered, there was a difference in the median F concentration according to manufacturer. Figure 2 shows the mean F concentrations of the 11 RTF infant milks manufactured by four companies for different stages of an infants growth. From the three different batch numbered samples of each milk analysed SMA from birth milk had the highest median F concentration, twice that of the Milupa

0.035 0.030 F concentration (g/ml) 0.025 0.020 0.015 0.010 0.005

From birth From 6 months Extra hungry from birth

Suggested age of consumption

Table 2 describes the F concentration of the 147 RTF foods and drinks by manufacturer-specied age group, based on age group for appropriate use stated by the manufacturer and each product being categorised once. A trend was observed which

0.000

Milupa

Cow and Gate

Heinz

SMA

Ready-to-feed infant milk

Fig. 2. The F concentrations of 11 ready-to-feed infant milks (lg ml) manufactured by four companies according to the different stages of an infants growth.

29

Maguire et al.
1.4 1.2 F Concentration (g/g) 1.0 0.8 0.6 0.4 0.2 0.0

Boots Cow and Ella's Heinz Hipp Mumtaz Organix Plum (n = 9) Gate kitchen (n = 16) (n = 16) (n = 3) (n = 6) (n = 5) (n = 21) (n = 3) Ready-to feed savoury meals

Fig. 3. The median and range of F concentrations (lg g) of 79 ready-to-feed savoury meals, by manufacturer.

was less than mixed meat products, which included other ingredients (such as vegetables) and which showed the highest median F concentration. Of the RTF infant foods drinks that were analysed, 94% contained less than 0.31 lgF g. Only 1% of the RTF products had an F concentration of more than 1.00 lgF g. However, it is not only the F concentrations of products but also variations in food drink intakes and dietary habits which determine F ingestion in infants, and therefore, F intake and excretion studies in infants do need to be an integral part of quantication and monitoring of F exposure in this age group.

and Heinz from birth milks and three times the median F concentration of the Cow & Gate equivalent milk.

RTF infant foods

When the current results were compared with those from other studies (Table 3), the F concentrations found for infant foods in the present study were higher than the mean F concentration of <0.3 lg g found for UK canned meat products, <0.5 lg g for jar meat products, and <0.2 lg g for canned and jar desserts (23). However, when compared with the other UK work by Vlachou and colleagues (24) who reported F concentration for 83 baby foods and 30 baby drinks purchased in Leeds, England, the values are similar or lower. In that study, of the RTF products, dairy products ranged from 0.02 to 0.18 lgF g, cereals from 0.04 to 0.70 lgF g, vegetable products from 0.03 to 0.48 lgF g, fruits from 0.03 to 0.07 lgF g, and desserts from 0.02 to 0.28 lgF g (24). Internationally, commercially available infant foods have shown a wide range of F concentrations as can be seen in Table 3(a), with the greatest variation seen for meat products. The study by Stamm and Kuo (29) found poultry products had a mean concentration of 1.82 lgF g, which the authors suggested was because of its processing involving mechanical deboning that can leave skin (which contains about 1 lgF g) and small particles of bone in the food. More recently, 206 RTF infant foods purchased in Iowa, US (30), were found to contain 1.058.38 lgF g for chicken, with meats (including poultry) showing the greatest variation in F concentration, ranging from 0.01 to 8.38 lgF g. In contrast, where meats do not form as large a component of an infants diet, the F concentrations of the representative food groups differ. The F concentrations of 11 RTF infant foods purchased in Japan, including sh dishes, showed a range from 0.46 to 2.94 lg g (31) with a sardine-based rice gruel having the greatest F concentration in the foods tested. More recently, Tomori et al. (32)

Savoury meals
The median and range of F concentrations for 79 RTF savoury meals manufactured by eight companies are shown in Fig. 3. Boots savoury meals showed the lowest median F concentration, while Mumtaz and Heinz meals had the highest median F concentrations. The greatest range of F concentrations for products was shown by Cow & Gate, who produced a broad range of products (n = 21) in a number of food and drink categories, followed by Hipp (n = 16), then Heinz (n = 16).

Discussion
In this study, 147 RTF products, 122 foods and 25 drinks, manufactured by a total of 12 companies were analysed for their F concentration. Although a convenience sample, the products were widely available in supermarket and pharmacies and represented the majority of commonly used RTF infant foods, drinks and milks currently available in the UK. The method of F analysis used to determine the F concentration of these foods and drinks was valid and provided a reliable estimate to within 0.001 lgF g. The F concentration of RTF infant milk formula as well as other infant drinks and foods analysed showed a wide variation. However, the range was not as wide as with foods and drinks requiring preparation before feeding (RPF), in which the nal F content as consumed is highly dependent on the F concentration of the water (or milk) used in reconstitution. The F content of RTF infant formula

30

Table 3. F content of RTF infant foods (lg g) and drinks(lg ml); studies carried out 19752004 for (a) Foods, (b) Drinks, (c) Formula milk

(a) Ready-to-feed (RTF) Foods (lg g) Country US US Canada UK UK Japan US Brazil Brazil Japan US UK 0.010.64 0.020.19 0.844.77 0.030.22 0.010.64 0.040.31 0.060.44 0.040.56 0.27.84 0.436.64 0.01 0.49 (fruits and deserts) 0.010.31 0.010.42 0.018.38 (1.058.38 for chicken) 0.010.63 0.0250.120 0.030.07 0.040.70 0.030.48 0.040.39 1.244.89 0.021.29 <0.10 <0.10 <0.10 0.541.29 0.631.17 0.282.08 0.320.95 n = 300 Number of samples not reported Fruits Cereals Vegetables Meats Mixed dishes Notes

Author (Date)

Wiatrowski et al. (1975) (38)

Stamm & Kuo (1977) (29)

Dabeka et al. (1982) (36) Walters et al. (1983) (23)

0.123.58 (strained) <0.10 (except chicken; 1.82) 0.053.93 <0.3 (canned) <0.5 (jar)

Vlachou et al. (1992) (24)

Nishijima et al. (1993) (31)

Heilman et al. (1997) (30)

n = 334 <0.2 lg g for canned and jar desserts n = 83; 0.020.18 lg g for dairy products 0.020.28 lg g for desserts n = 11; 0.462.94lg g for infant foods incl. sh n = 206

Buzalaf et al. (2002) (33)

Buzalaf et al. (2004) (34)

Tomori et al. (2004) (32)

n = 10; 0.061.28lg g for dairy products n = 7; 0.311.8 lg g for infant biscuits n = 16 0.070.45 (excl meat) n = 116; 0.060.71 lg g for pasta n = 122; 0.040.40 lg g for baked goods; 0.030.38 lg g for desserts

Chow et al. (2005) (35) Maguire et al. (present study)

0.020.18 (meat and sh) 0.012.54 0.041.20

b) Ready-to-Feed Drinks (lgF ml) Country US UK Brazil Japan US UK 0.010.26 0.010.30 0.140.18 0.010.80 0.050.15 0.022.80 Range Mean 0.65 Notes

Author (Date)

Kiritsy et al. (1996) (40,41)

Fluoride content of infant food and drinks

Vlachou et al. (1992) (24) Buzalaf et al. (2004) (34) Tomori et al. (2004) (32) Chow et al. (2005) (35) Maguire et al. (present study)

0.08

n = 532; White grape juice (1.40), red grape juice (0.47), tea (1.41), prune juice (0.79) n = 24 (n = 8) Soya beverages (n = 3) Fruit juices Juices and puddings (n = 8), median = 0.07. Highest F concentration found in a peach and pear juice

31

Maguire et al.

analysed 68 samples of commercial infant foods and drinks in Japan, 19 (16 foods, three drinks) of which were RTF products. Of these retort-packed products, the F concentrations ranged from 0.14 to 0.18 lg g for fruit juices, 0.040.56 lg g for vegetables, 0.020.18 lg g for meat & sh and 0.06 0.44 lg g for cereals. In general, in the present study, of the eight food and drink categories for RTF products, the baked goods had the highest F content (median 0.196; range 0.0400.397 lg g), which would appear to reect the water content of dough, pastry and the cooked rice which formed the basis of this food group. The study by Buzalaf et al. (33) of the South American infant and weaning diet found that infant cereals had a much higher F concentration range of 0.43 to 6.64 lg g, while analysis of another six infant cereals and seven biscuits in Brazil showed a range of 0.2 to 7.8 and 0.3 to 11.8 lgF g, respectively, with nonlled calcium-rich biscuits having the greatest F concentration (34). The analyses completed by Chow et al. also showed a wide range of F concentrations for cereals (0.84 4.77 lg g) in Georgia, US (35). The range of 0.030 0.221 lgF g for infant cereals obtained in the present study is also smaller than the range for infant cereals on the Canadian market (1.24 to 4.89 lgF g) found in the study by Dabeka et al. (36), while lower concentrations have also been reported for infant cereals in Iowa, US (30) and Japan (32), respectively (Table 3a). When the six RTF breakfast cereals were compared with 27 cereals which required preparation with milk or water Zohoori et al. (37), as Table 4 shows, for the six RTF breakfast cereals, the median (range) F concentration was 0.110 (0.0300.221) lg g, while for the 10 RPF breakfast cereals that recommended the use of milk for re-constitution, it was 0.083 (0.0530.10) lg g. In contrast, for breakfast cereals that required preparation with water, the median (range) F concentrations were higher; 0.161 (0.1100.340) lg g when low F water (0.13 mgF L) was used and 0.908 (0.4881.230) lg g when uoridated (0.90 mgF L) water was used. Therefore, although RTF products may have a more limited shelf life or increased storage problems because of their ready to feed rather than their Requiring Preparation prior to Feeding (RPF) properties, they may have an important use where an alternative to breastfeeding is required in areas where F-free or low F waters of a suitable quality are not readily available for use with RPF products.

Mean

0.901.00 0.250.32 0.200.86 0.100.36 (processed in non-F area) 0.231.01 (processed in F area)

0.29

Country

Canada Canada US US

c) Ready-to-Feed Infant formula (lgF ml)

Johnson & Bawden (1987) (46)

US Van Winkle et al. (1995) (44)

Table 3. Continued

32

Maguire et al. (present study)

Farkas & Farkas (1974) (42) Wiatrowski et al. (1975) (38) Tinanoff & Muller (1978) (43) Singer & Ophaug (1979) (39)

Author (Date)

UK

US

0.040.55 (cows milk) 0.170.38 (soya-based milk) 0.010.03

Range

0.050.38

0.17 0.30 0.02

n=2 n=4 n=8 n = 16 Higher F concentrations found in the soya-based milks n = 51 (15 products purchased in up to 7 states) n = 16 n=5 Median = 0.02

Notes

Fluoride content of infant food and drinks Table 4. The median (range) F concentrations (lg g) of ready-to-feed infant breakfast cereals (n = 6) and infant breakfast cereals requiring preparation before feeding by adding either milk (n = 10) or water (n = 17) according to the manufacturers instructions, by manufacturer (from the study by Zohoori et al. (37); submitted to Community Dent Oral Epidemiol) RTF Median (range) [n] Bebivita Boots Cow & Gate Heinz Hipp Organix Total
a

Prepared with milka Median (range) [n] 0.067 [3] 0.087 [3] 0.100 [1] 0.081 [3] (0.0530.072) (0.0840.093) ()) (0.0790.091)

Prepared with NF waterb Median (range) [n] 0.223 [2] 0.161 [1] 0.315 [2] 0.160 [5] 0.211 [3] 0.132 [4] 0.161 [17] (0.1980.248) ()) (0.2900.340) (0.1100.220) (0.2060.250) (0.1280.151) (0.1100.340)

Prepared with F waterb Median (range) [n] 0.845 [2] 1.059 [1] 1.030 [2] 0.770 [5] 0.890 [3] 1.027 [4] 0.908 [17] (0.7810.908) ()) (0.8301.230) (0.7101.010) (0.4881.040) (0.9891.203) (0.4881.230)

0.100 [2] 0.221 [1] 0.070 [2] 0.110 [1] 0.110 [6] (0.0480.152) ()) (0.0300.110) ()) (0.0300.221)

0.083 (0.0530.100) [10]

Ready-to-feed infant milk; SMA gold (0.030 lgF ml). Same 17 products, re-constituted with nonuoridated (0.13 mgF l) or uoridated (0.90 mgF l) water.

In the present study, it was the Savoury meals group (equivalent to the mixed foods shown in Table 3a) that showed the greatest range of F content (from 0.040 to 1.200 lg g), with the upper end of this range being accounted for primarily by the 14 meat products (median F concentration 0.147 lg g) and in particular three products (pork-, lamb- and beef-based) with a F content of >0.5 lg g. This contrasts with a number of studies of infant foods containing chicken which have reported a higher F concentration because of mechanical deboning of chicken leaving residual bone particles in the food (30, 38, 39). The differences seen between the mean F concentration of savoury meals according to manufacturer could also be due to the F concentration of water used for processing. It should also be noted that the numbers of savoury meal products tested within the subgroups were not equal for all the manufacturers. For example, only three savoury products were sourced for the Mumtaz brand, compared with 19 Cow & Gate products. As Table 3(a) shows, analysis of RTF infant mixed foods in Illinois, US (38) and Iowa, US (30), showed a range of F concentrations considerably higher than the range (median) of 0.0401.02 (0.112) lgF g estimated in this study for RTF mixed foods. This may also be due to the F concentration of the water used for processing, as a larger proportion of North America receives a uoridated water supply compared with the UK.

Reporting values for F concentrations to compare foods from different countries, areas and manufacturers can be difcult when infant foods and drinks are tested by different investigators. This is one reason why attempts to standardise methods for collection, preparation and analysis of food and drinks for F content are very important and welcome and to be encouraged in this area of F research (27).

RTF infant drinks

The range of F concentrations for RTF infant juices in the present study was 0.050.15 lg g with the highest F concentration found in a pear and peach juice. This was similar to the values found for RTF products in other studies (Table 3b), especially where the manufacturing plant was sited in a nonF area. In contrast, for those drinks manufactured in F areas, the F content tends to be much higher, for both infant drinks (39, 40) and the 532 infant juices and juice-avoured drinks identied and analysed by Kiritsy et al. (41) in a large long-term prospective early childhood F exposure study in Iowa, US. In that study, a wide range of F concentrations from 0.02 to 2.80 lg ml with a mean of 0.65 lg ml was reported with white grape juice having a relatively high F concentration (1.40 lg ml), similar to tea (1.41 lg ml), but higher than prune juice (0.79 lg ml) and red grape juice (0.47 lg ml). As Table 3(b) also shows, in the UK, a similar range of F concentrations from 0.01 to 0.26 lg ml

33

Maguire et al.

for 24 RTF infant drinks was reported by Vlachou et al. (24). More recent analysis of eight soya beverages in Brazil also showed a similar range of F concentrations, from 0.1 to 0.3 lg ml (34), while Tomori and co-workers (32) found fruit juices (n = 3) in Japan contained 0.140.18 lgF ml.

RTF infant formula milks

The 11 RTF infant formula milks analysed for F content in the present study showed a median (range) F concentration of 0.016 (0.0090.030) lg ml. As Table 3(c) shows, this is lower than earlier reported concentrations in RTF milks processed in nonuoridated and uoridated areas of North America (38, 39, 42, 43) as well as more recent analyses carried out in Iowa, US (44). The distinction between earlier and later studies is important, as it was the earlier work, primarily in the 1970s which drew attention to the potential for higher F exposure from infant milks. As a result, manufacturers of infant formula milks agreed to limit the F content of the water used in processing to <0.15 mgF l (15, 43, 45, 46). As a result of this development, more recent studies have shown a more consistent pattern of lower F concentrations for the newer RTF products, achieved through changes in manufacturing practice; in particular, the close monitoring and evaluation of the chemical quality of waters used in infant food processing and formulation (45). Despite breastfeeding being the preferred feeding choice for infants, some may be fed exclusively on formula milk for the rst six months of life. According to the feeding guidelines provided on packaging by RTF infant formula manufacturers, an infant up to the age of 6 months is recommended to drink approximately 150 ml milk kg bw per day (9). Based on the median F concentration of RTF infant formulas of 0.016 lg ml, found in the present study, the F intake for infants using these products would be 0.0024 mg kg bw per day. These values are well below the F intake predicted for infants up to the age of 4 months in Ireland (0.11 and 0.14 mg kg bw per day) (47, 48) and the safe F intake threshold, to avoid the risk of dental uorosis, of 0.22 mgF kg bw per day in infants under 6 months suggested by the Committee on Medical Aspects of Foods (49) as well as the tolerable upper intake level of 0.1 mg kg bw per day dened by the European Food Safety Agency (EFSA) for 18-year-olds (10). The wide variation in F concentrations of RTF infant foods and drinks between different manufacturers observed in this study, particularly with

the savoury meals and milks concentrations, may be attributed to different methods of processing, different ingredients and the areas of origin of the ingredients (including water). The F concentration of water used to prepare infant foods and drinks has an important effect on their F concentration as consumed in both RTF and re-constituted (RPF) products, either included as part of the processing (RTFs and RPFs) or introduced by the consumer when the meal is re-constituted during home preparation (RPFs). The choice of RTF products for infant feeding, while more expensive, appears to provide a more consistent F concentration open to less variation between manufacturers and also at the point of consumption, being ready to feed rather than requiring re-constitution, and professional bodies are starting to address this issue in their recommendations. A recent report (50) from the American Dental Associations Council on Scientic Affairs advised dental practitioners to suggest RTF formula or powdered or liquid concentrate formula reconstituted with water that is either F-free or contains only low concentrations of F when advising parents and caregivers who may be concerned about the potential for increasing a childs risk of developing dental uorosis. An infant could ingest F in higher than desirable amounts of F from RTF foods if they eat large quantities of foods containing greater than average F concentrations, e.g. certain meat-based savoury meals; however, this is much less likely for RTF foods than for RPF foods prepared with F water. While the consumer can control the quantities consumed, the manufacturer has the major control over the F content of RTF infant foods and drinks and therefore has an important role, either at the production stage or through provision of appropriate preparation instructions on the product, to help avoid over-exposure to F. It is important that parents receive appropriate information on the F content of infant foods and drinks, and advice about when RTF foods and drinks may be appropriate or useful in their infants diet. This information would be most appropriately disseminated by health professionals supported by appropriate manufacturers instructions and labelling of products. In view of recent studies and recommendations of expert bodies, the relevant guidelines and advice for infant feeding practices need to be reviewed alongside the current labelling, including manufacturers instructions, provided on infant food products. With this in mind, further

34

Fluoride content of infant food and drinks

research on actual F intake in infants conducted through observational studies would be useful to underpin the evidence base for any future renements of infant feeding guidelines with respect to F exposure. In summary, the F concentrations of UKmarketed RTF infant foods and drinks analysed in this study ranged from a median (range) of 0.016 (0.0090.03) lgF ml for infant formula milks to 0.196 (0.0400.397) lgF g for baked goods and 0.147(0.0431.200) lgF g for meats. For most products, these concentrations do not appear to be sufciently high to be a risk factor for dental uorosis, if consumption is within the normal limits of the diet recommended for infants and young children. However, they can make a substantial contribution to the total daily F intake of an infant. Knowledge of the F intake of infants and children is essential for optimising the benecial use of F to achieve oral health and also to facilitate decision-making on community-based prevention methods including water uoridation. Further research is needed to determine the F intake of infants living in both uoridated and nonuoridated communities.

Acknowledgements
This study was supported by a grant from the Organix Foundation.

References
1. McDonagh M, Whiting P, Bradley M, Cooper J, Sutton A, Chestnutt I et al. A systematic review of public water uoridation. York: NHS Centre for Reviews and Dissemination, University of York; 2000. 2. Singh K, Spencer A, Armeld J. Relative effects of pre- and posteruption water uoride on caries experience of permanent rst molars. J Public Health Dent 2003;63:119. 3. Pendrys DG. Risk of enamel uorosis in nonuoridated and optimally uoridated populations: considerations for the dental professional. J Am Dent Assoc 2000;131:74655. 4. Tabari ED, Ellwood RP, Rugg-Gunn AJ, Evans DJ, Davies RM. Dental uorosis in permanent incisor teeth in relation to water uoridation, social deprivation and toothpaste use in infancy. Br Dent J 2000;189:21620. 5. Tsutsui A, Yagi M, Horowitz AM. The prevalence of dental caries and uorosis in Japanese communities with up to 1.4 ppm of naturally occurring uoride. J Public Health Dent 2000;60:14753. 6. Whitford GM. Intake and metabolism of uoride. Adv Dent Res 1994;8:514.

7. Burt BA. The changing patterns of systemic uoride intake. J Dent Res 1992;71:122837. 8. Department of Health. An oral health strategy for England. London: Department of Health; 1994. 9. Department of Health. Infant feeding recommendations. London: Department of Health; 2003. 10. European Food Safety Authority (EFSA). Opinion of the scientic panel on dietetic products, nutrition and allergies on a request from the commission related to the tolerable upper intake level of uoride. Eur Food Safety Authority J 2005;192:165. 11. US Environmental Protection Agency (USEPA). Fluorine (soluble uoride). Integrated risk information system. Cincinnati, Ohio: US Environmental Protection Agency; 2003. 12. Erdal S, Buchanan SN. A quantitative look at uorosis, uoride exposure, and intake in children using a health risk assessment approach. Environ Health Perspect 2005;113:1117. 13. Marthaler TM. Monitoring of renal uoride excretion in community preventive programmes on oral health. Geneva: World Health Organization; 1999. 14. World Health Organization. Fluorides and oral health. Report of a WHO expert committee on oral health status and uoride use. Tech Rep Ser 1994;846:137. 15. Ekstrand J. Fluoride intake in early infancy. J Nutr 1989;119:185660. 16. Levy SM, Warren JJ, Broftt B. Patterns of uoride intake from 36 to 72 months of age. J Public Health Dent 2003;63:21120. 17. Levy SM, Warren JJ, Davis CS, Kirchner HL, Kanellis MJ, Wefel JS. Patterns of uoride intake from birth to 36 months. J Public Health Dent 2001;61:707. 18. Research and Markets. Baby foods market assessment, 2009. Available from: http://www.researchandmarkets.com/reports/3492 [accessed 1st June 2009]. 19. Holland B, Welch A, Unwin I, Buss D, Paul A, Southgate D. McCance and Widdowsons; The composition of foods. London: The Royal Society of Chemistry and Ministry of Agriculture, Fisheries and Food; 1992. 20. U.S. Department of Agriculture (USDA). National uoride database of selected beverages and foods. Prepared by Nutrient Data Laboratory, Beltsville Human Nutrition Research Center, Beltsville, Maryland. Agricultural Research Service and U.S. Department of Agriculture, 2005. 21. Zohouri FV, Maguire A, Moynihan PJ. Sources of dietary uoride intake in 6-7-year-old English children receiving optimally, sub-optimally, and non-uoridated water. J Public Health Dent 2006;66: 22734. 22. Maguire A, Zohouri FV, Hindmarch PN, Hatts J, Moynihan PJ. Fluoride intake and urinary excretion in 6- to 7-year-old children living in optimally, suboptimally and non-uoridated areas. Commun Dent Oral Epidemiol 2007;35:47988. 23. Walters CB, Sherlock JC, Evans WH, Read JI. Dietary intake of uoride in the United Kingdom and uoride content of some foodstuffs. J Sci Food Agric 1983;34:5238. 24. Vlachou A, Drummond BK, Curzon ME. Fluoride concentrations of infant foods and drinks in the united kingdom. Caries Res 1992;26:2932.

35

Maguire et al. 25. Taves DR. Separation of uoride by rapid diffusion using hexamethyldisiloxane. Talanta 1968;15:969 74. 26. Zohouri FV, Rugg-Gunn AJ. Fluoride concentration in foods from Iran. Int J Food Sci Nutr 1999;50:265 74. nez-Mier EA, Cury JA, Heilman JR, Katz BP, 27. Mart Levy SM, Li Y et al. Development of gold standard ion-selective electrode-based methods for uoride analysis. Caries Res 2011;45:312. 28. Zohouri FV, Maguire A, Moynihan PJ. Fluoride content of still bottled waters available in the northeast of England, Uk. Br Dent J 2003;195:5158. 29. Stamm JW, Kuo HC. Fluoride concentration in prepared infant foods. J Dent Res 1977;56:1. 30. Heilman JR, Kiritsy MC, Levy SM, Wefel JS. Fluoride concentrations of infant foods. J Am Dent Assoc 1997;128:85763. 31. Nishijima MT, Koga H, Maki Y, Takaesu Y. A comparison of daily uoride intakes from food samples in Japan and Brazil. Bull Tokyo Dent Coll 1993;34:4350. 32. Tomori T, Koga H, Maki Y, Takaesu Y. Fluoride analysis of foods for infants and estimation of daily uoride intake. Bull Tokyo Dent Coll 2004;45: 1932. 33. Buzalaf MAR, Granjeiro JM, Duarte JL, Taga MLdL. Fluoride content of infant foods in Brazil and risk of dental uorosis. J Dent Child 2002;69:196200. 34. Buzalaf MA, de Almeida BS, Cardoso VE, Olympio KP, Furlani Tde A. Total and acid-soluble uoride content of infant cereals, beverages and biscuits from Brazil. Food Addit Contam 2004;21:2105. 35. Chow B, Chow D, Whitford GM. Fluoride in infant foods: concentrations and associated intakes. J Dent Res 2005;84(Spec Issue A):2387. 36. Dabeka R, McKenzie A, Conacher H, Kirkpatrick D. Determination of uoride in Canadian infant food and calculation of uoride intakes by infants. Can J Public Health 1982;73:18891. 37. Zohoori FV, Moynihan PJ, Omid N, Abuhaloob L, Maguire A. Fluoride concentrations of infant foods and drinks in the United Kingdom. Caries Res 2009; 43:204. 38. Wiatrowski E, Kramer L, Osis D, Spencer H. Dietary uoride intake of infants. Pediatrics 1975;55:51722. 39. Singer L, Ophaug R. Total uoride intake of infants. Pediatrics 1979;63:4606. 40. Silva M, Reynolds EC. Fluoride content of infant formulae in Australia. Aust Dent J 1996;41:3742. 41. Kiritsy MC, Levy SM, Warren JJ, Guha-Chowdhury N, Heilman JR, Marshall T. Assessing uoride concentrations of juices and juice-avored drinks. J Am Dent Assoc 1996;127:895902. 42. Farkas CS, Farkas EJ. Potential effect of food processing on the uoride content of infant foods. Sci Total Environ 1974;2:399405. 43. Tinanoff N, Mueller B. Fluoride content in milk and formula for infants. J Dent Child 1978;45:535. 44. Van Winkle S, Levy SM, Kiritsy MC, Heilman JR, Wefel JS, Marshall T. Water and formula uoride concentrations: signicance for infants fed formula. Pediatr Dent 1995;17:30510. 45. Feigal RJ. Recent modications in the use of uorides for children. Northwest Dent 1983;62:1921. 46. Johnson J Jr, Bawden JW. The uoride content of infant formulas available in 1985. Pediatr Dent 1987;9:337. 47. Food Safety Authority of Ireland (FSAI). Risk assessment of the uoride intake of Irish infants under 4 months of age as a result of the consumption of infant formula constituted with uoridated water. Dublin: FSAI; 2002. 48. Anderson K, Pratt I, Ryan M, Flynn A. A probabilistic estimation of uoride intake by infants up to the age of 4 months from infant formula reconstituted with tap water in the uoridated regions of Ireland. Caries Res 2004;38:4219. 49. Committee on the Medical aspects of Food Policy (COMA). Weaning and the weaning diet. Report on Health and Social Subjects. London: Department of Health; 1994. 50. Berg J, Gerweck C, Hujoel PP, King R, Krol DM, Kumar J et al. Evidence-based clinical recommendations regarding uoride intake from reconstituted infant formula and enamel uorosis: a report of the American Dental Assoc Council on Scientic Affairs. J Am Dent Assoc 2011;142:7987.

36

This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material.