Community Dent Oral Epidemiol 2014; 42; 224–233 All rights reserved
Ó 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Fluoride intake from fluids and urinary fluoride excretion by young children in Kuwait: a nonfluoridated community
Enosakhare S. Akpata1, Jawad Behbehani1, Jaber Akbar1, Lukman Thalib2 and Olusegun Mojiminiyi3 1 Department of Restorative Sciences, Kuwait University Safat, Kuwait, 2Department of Community Medicine, Kuwait University Safat, Kuwait, 3Department of Pathology, Kuwait University Safat, Kuwait
Akpata ES, Behbehani J, Akbar J, Thalib L, Mojiminiyi O. Fluoride intake from fluids and urinary fluoride excretion by young children in Kuwait: a nonfluoridated community. Community Dent Oral Epidemiol 2014; 42: 224–233. © 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Abstract – Objectives: To determine the pattern of fluid consumption, fluoride intake from the fluids and urinary fluoride excretion by children aged 1– 9 years in Kuwait, a nonfluoridated community. Methods: Using the cluster sampling technique, children aged 1–9 years were chosen from 2000 randomly selected households in Kuwait. Questionnaires were then administered to their mothers to determine the children’s daily fluid intake. Fluoride concentrations in tap water as well as all brands of bottled water and beverages consumed by the children were measured, using the fluoride ion-specific electrode. Fluoride excretion was determined in 400 randomly selected children, based on fluoride/creatinine ratio. Results: The mean daily fluid consumption by the children was high, being 1115–1545 ml. About 40% of the fluid intake was plain (tap and bottled) water and approximately 10% of the children drank bottled water exclusively. Fluoride concentration in tap water was low (0.04 SD 0.02 ppm), but was higher in bottled water (0.28 SD 0.40 ppm). Mean daily fluoride ingestion from fluids was 0.013–0.018 mg/kg body weight (bw). Even after allowing for fluoride ingestion from other sources, mean daily fluoride ingestion was still below 0.1 mg/kg bw set by the United States of America Institute of Medicine as the lowest-observed-adverse-effect level for moderate enamel fluorosis in children aged up to 8 years. Furthermore, the mean daily urinary fluoride excretion of 128–220 lg was below the provisional standard of 360–480 lg for optimal fluoride usage by children aged 3–5 years. Conclusion: Fluoride ingestion from fluids and urinary fluoride excretion by the children were below the recommendations for optimal fluoride usage. Thus, there is room for an upward adjustment of fluoride level in public drinking water supplies in Kuwait, as a caries preventive measure.
Exposure to fluoride from drinking water is a proven public health measure against dental caries (1–4). However, excessive fluoride ingestion during the period of tooth mineralization causes dental fluorosis (5–7). Appropriate fluoride concentration in drinking water varies in different parts of the world, depending on the climatic conditions. Galagan and Vermillion (8) suggested a formula for calculating appropriate fluoride level in drinking doi: 10.1111/cdoe.12081
Key words: cariology; epidemiology; fluoride; prevention; public health Enosakhare S. Akpata, Faculty of Dentistry, Kuwait University, P.O. Box 24923, Safat 13110, Kuwait Tel.: 965 2 498 6741 Fax: 965 2 498 6741 e-mail: [email protected] Submitted 29 November 2012; accepted 21 September 2013
water, based on ambient temperature. Thus, the appropriate fluoride concentration of approximately 0.7 ppm has been suggested for countries with a mean maximum annual ambient temperature above 27°C. Nevertheless, severe dental fluorosis has been observed in some tropical communities exposed to apparently appropriate fluoride level in drinking water. For example, in Saudi Arabia, severe dental
224
Fluoride ingestion and excretion in Kuwait
fluorosis was observed in some children exposed to well drinking water containing 0.5–0.8 ppm (9). Similar observations were reported in Senegal (10) and Sudan (11). It has been suggested that the high vulnerability to dental fluorosis in these tropical countries may be due to increased water consumption, and consequently excessive fluoride ingestion. Besides, Sohn et al. (12) recently showed that fluoride ingestion is related to fluid consumption, the patterns of which is undergoing changes in different parts of the world (13, 14). Other sources of fluoride ingestion include sea foods (15), toothpastes (16), infant formulas, and fluoride supplements (17, 18). Caries experience has been reported to be high among children in Kuwait and may be on the increase. A review of cross-sectional studies carried out on the occurrence of dental caries in the country between 1982 and 1994 showed a trend toward an increase in caries experience in the primary and permanent dentitions (19). Moreover, Vigild et al. (20) reported that caries had increased among children in Kuwait between 1982 and 1993. In a more recent national survey, Al-Mutawa et al. (21) reported caries prevalence to be 81.4% and 75.6% among children aged 5 and 6 years, respectively, and corresponding decayed and filled primary teeth (dft)/decayed and filled surfaces (dfs) to be 4.6/9.7 and 4.6/9.9, respectively. In contrast, caries prevalence among ethnic Danish children was reported to be 53% at age 7 years, while mean dmft was 3.5 (22). Fluoridation of public water supplies was discontinued in Kuwait in 1980; in view of the high caries experience among children in the country, an upward adjustment of the fluoride level in drinking water might be a viable caries preventive measure. If this were to be carried out, knowledge of fluoride ingestion from water and other fluids, as well as urinary fluoride excretion would be desirable, especially as these parameters appear to be unknown in Arab children during the period of tooth mineralization. The aim of this work, therefore, was to determine the pattern of fluid consumption, fluoride intake from ingested fluids, and daily urinary fluoride excretion by children aged 1–9 years in Kuwait, a nonfluoridated Arab community.
Subjects and methods Subjects Using the cluster sampling technique, children aged 1–9 years were chosen from about 2000
households from all the six governorates in Kuwait. Each of the governorates in the State of Kuwait is divided into suburbs or localities, and each suburb into blocks. Each block contains a certain number of streets. From each governorate, three suburbs were randomly selected, and from each suburb, three streets randomly chosen. Alternate houses on each selected street were then visited. The number of children included in the sample was calculated proportionally from the population of each governorate. The total number of children sampled from Ahmadi, Capital, Farwaniya, Hawalli, Jahra, and Mubarak Al-Kabeer governorates were 333, 153, 270, 756, 243, and 261, respectively, making a total of 2016 children sampled during each season (winter and summer). However, the mean number of mothers interviewed during each of the two seasons was 2022. The sample size was based on previous studies (13, 23, 24) to ensure a and b errors of 5% in the measurement of daily fluid ingestion (0.1 l), proportion (5%) of fluid intake that was plain water and fluoride ingestion (0.001 mg). Only one child was selected from each household. The children were in good general health and were not on any medication or fluoride supplements. If there was more than one child of the desired age in the selected household, one of them was chosen by ballot. If the mother in the household refused to participate or there was no child of the desired age, the interviewer proceeded to the next household. The refusal rate was approximately 20%. Approval for the research was obtained from the Ethics Committee of the Health Sciences Center, Kuwait University. The purpose of the research was explained to the parents of the selected children, and they gave their consent by signing the consent form. Demographic information was obtained from the children’s parents.
Daily fluid consumption and oral hygiene practices Prior to the study, four fieldworkers, who had previous experience with household surveys, were trained in the use of the questionnaire designed for the present study. Each of them then visited five households in different governorates in Kuwait. The research team discussed with them their findings and resolved potential problems. The pretested questionnaire was administered to mothers of the selected children by the trained interviewers. The questionnaire inquired about the types and quantities of all fluids ingested by the
225
Akpata et al.
children within the previous 24 hours, that is, the fluid intake before breakfast, at breakfast, between breakfast and lunch, at lunch, between lunch and dinner, at dinner and after dinner before going to bed. Household measures and color photographs of various sizes of bottled water and beverages were used to estimate volumes. Administration of the questionnaire was for 3 days in summer and another 3 days in winter. This was repeated for the same child after an interval of at least 3 days during each of the seasons. The results for the first and second interviews were averaged for each of the seasons. The questionnaire also inquired about the children’s oral hygiene practices, including the type, amount, and frequency of fluoride toothpaste use. The size of toothpaste was measured in terms of toothbrush lengths, illustrated by color photographs.
Measurement of weights and heights The children’s weights were measured with calibrated bathroom scales (Terraillon, Croissy-SurSeine Cedex, France) to the nearest 0.5 kilogram. While measuring their weights, the children put on light clothing and were barefooted. With the children barefooted, and standing against a vertical wall, their heights were measured with a tape rule to the nearest 1 cm, with the child’s heels, back, and occiput touching a vertical wall, and a horizontal bar lowered to touch the vertex. From the measurements, body mass index (BMI) was calculated. BMI = weight (kg)/height (m2).
Fluoride concentrations in drinking water and beverages An experienced technician from the Faculty of Science, Kuwait University, carried out fluoride analysis of the samples. Preliminary analyses were carried out on 15 samples of plain water, beverages, and urine. Duplicates of the samples, to which the technician was blinded, were re-analyzed for fluoride two days later. Correlation plot showed very high correlation (r = 0.9, P < 0.0001) between the results of the two measurements, indicating high reproducibility. Tap water samples collected from households, different commercial brands of bottled water, cold, and hot beverages (including tea) consumed by the children were analyzed in duplicate for fluoride content, using the Fluoride Ion Selective Electrode, ISE (Model 960900; Thermo Electron Corporation,
226
MA, USA) in conjunction with an ISE Meter (Orion Model 720A; Thermo Electron Corporation, Beverly, MA, USA). Twenty-five milliliters of each sample was mixed with equal volume of Total Ionic Strength Adjusting Buffer II (TISAB II), before analysis. Fluoride standards ranging from 0.01 to 1 mg/l were used to calibrate the measurements. The tap water (about 200 ml) was obtained in polythene bottles from four taps within households in each of the six governorates in Kuwait. The polythene bottles were rinsed three times with the tap water before sample collection. Samples of bottled water and beverages were from the various brands bought from supermarkets and grocery stores in Kuwait. Tea infusions for the analysis were prepared by adding one tea bag to 100 ml of boiled de-ionized water, and the tea brewed for 2 minutes. Powdered beverages were reconstituted according to the manufacturers’ instructions/recommendation, using de-ionized water. Fluoride ingestion from tap water was calculated by multiplying fluoride concentration by the volume of water consumed. In the case of other fluids, fluoride ingestion was calculated for each child by multiplying the volume of the fluid consumed by the fluoride concentration in that particular brand of fluid.
Estimation of urinary fluoride excretion Early morning spot urine sample was obtained from the child at every fifth household visited. The fieldworker gave the mother 60-ml polythene bottle at the first visit to the household and collected the urine sample at the second visit. Because of the practical difficulty in collecting 24-hour urine samples from many children in a study of this nature, we estimated 24-hour urinary fluoride excretion from the fluoride/creatinine ratio of the early morning spot urine samples. The samples were stored at 20°C, until analyzed in triplicates, using the ISE after adding TISAB II. Urinary creatinine concentrations of the spot urine samples were measured by the kinetic Jaffe reaction (25), using the Beckman Synchron LX20 automated analyzer (Beckman Corporation, Brea, CA, USA). From the concentrations of urinary fluoride and urinary creatinine measurements, the fluoride/creatinine ratio was calculated. The subject’s 24-hour urinary fluoride excretion was then calculated by multiplying the ratio by the standard creatinine value [15 mg/l/kg body weight (bw)/ day] and the child’s bw (26).
Fluoride ingestion and excretion in Kuwait
Statistical analysis
the mothers interviewed was rather high: about 41% had tertiary/university education, while only about 10% had primary or no formal education. Over 98% earned at least the recommended minimum salary (about $1500 per month) for families in Kuwait. Only about 3% of the children were obese (BMI ≥ 25). The mean maximum annual temperature during the summer months was 44°C and in winter 22°C.
Descriptive statistics were presented as frequency, percentages, means, and standard deviations. ANOVA was used to determine the statistical significance of the differences between fluid consumption, fluoride ingestion, and excretion among children of different sexes, age groups, nationality, and BMI; Multiple logistic regression analysis was used to determine factors associated with high fluoride ingestion (cutoff point set at 0.02 mg/kg bw/day, i.e. the 75th percentile of the children’s fluoride ingestion). The data management and statistical analysis were carried out using the statistical software SPSS-PC version 18.0 (IBM, New York, NY, USA).
Fluid consumption Mean daily fluid consumption among children in Kuwait was relatively high (Table 1), being 1129 302, 1372 321, and 1599 311 ml in summer and 1149 231, 1331 271, and 1538 282 ml in winter at ages 1–3, 4–6, and 7–9 years, respectively. ANOVA showed the age and seasonal differences to be statistically significant (P < 0.05). The only exceptions were the consumption of bottled water in which there were no significant age differences, and carbonated drinks in which there were no seasonal differences. In general, standard deviations were rather high, indicating the great variability in fluid consumption. Plain water constituted 41% of fluid intake by 1–9-year-old children in Kuwait in summer, of which about 6% was bottled water, the remainder being tap water. In winter, however, plain water constituted 36% of fluid intake, of which about 5% was bottled water (Table 1). About 11% of the children drank bottled water exclusively, while 2% never drank it. Most of the children (87%) drank bottled water occasionally.
Results Demographic characteristics The questionnaire study was carried out in 2009– 2011, and 2018 mothers of children aged 1–9 years were interviewed in summer and 2026 in winter. The ratio of boys to girls was 52:48, and their distribution in the age groups 1–3, 4–6, and 7–9 years was reasonably even. About 95% of the children were Arabs, and about 60% Kuwaitis (Kuwaitis are Arabs and non-Kuwaiti Arabs include those from other countries in the Middle East, such as Saudi Arabia, Egypt, United Arab Emirates and Qatar). Among non-Arabs in Kuwait are Indians, Pakistanis and Westerners). The mean bws of the children aged 1–3, 4–6, and 7–9 years were 13.2, 20.7, and 30.6 kg, respectively. The level of education of
Table 1. Frequency distribution of daily fluid consumption (ml) among children in Kuwait, by age groups and season*, based on questionnaire interviews Summer
Winter
Age groups (years)
Age groups (years)
1–3
4–6
7–9
1–3
4–6
7–9
N = 561
N = 704
N = 753
N = 613
N = 726
N = 687
Fluids
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Total fluid Plain water Tap water Bottled water Carbonated beverages Fruit juices and milk Hot beverage (tea) Hot beverages (other than tea)
1129 430 335 95 43 450 10 195
302 186 239 169 62 213 34 215
1372 584 515 69 116 543 58 69
321 226 264 178 94 167 79 84
1599 672 586 86 183 572 118 53
311 238 293 211 119 186 102 82
1149 367 300 66 48 410 16 309
231 135 187 137 64 211 42 249
1331 494 422 71 120 510 80 129
271 171 236 166 98 164 92 145
1538 581 512 69 169 527 155 106
282 198 264 181 111 163 109 127
*Two-way ANOVA (age groups and seasons), P < 0.05, except for carbonated drinks (P = 0.475), and bottled water (P = 0.278).
227
Akpata et al.
Children whose mothers had higher formal educational background and income consumed significantly less fluid (P < 0.05). However, there was no significant difference in fluid consumption between the Kuwaitis and non-Kuwaitis, irrespective of the type of fluid. There was a significant association between beverage ingestion and obesity although the correlation was weak (r = 0.16, P < 0.001).
Fluoride concentration in fluids The children consumed many brands of bottled water and beverages: 15 brands of bottled water, 30 of juices, 26 soft drinks of which eight were carbonated, eight of hot beverages including tea, and 15 of milk products, most of which were simply milk flavored, making a total of 94 brands of bottled water and beverages. Table 2 summarizes the fluoride levels in bottled water, liquid, and powdered beverages. It can be seen that over 66% of the bottled water and beverages had fluoride concentrations of 0.01–0.099 ppm. None of the brands of bottled water had fluoride level above 1.10 ppm of fluoride, while
Ó 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Fluoride intake from fluids and urinary fluoride excretion by young children in Kuwait: a nonfluoridated community
Enosakhare S. Akpata1, Jawad Behbehani1, Jaber Akbar1, Lukman Thalib2 and Olusegun Mojiminiyi3 1 Department of Restorative Sciences, Kuwait University Safat, Kuwait, 2Department of Community Medicine, Kuwait University Safat, Kuwait, 3Department of Pathology, Kuwait University Safat, Kuwait
Akpata ES, Behbehani J, Akbar J, Thalib L, Mojiminiyi O. Fluoride intake from fluids and urinary fluoride excretion by young children in Kuwait: a nonfluoridated community. Community Dent Oral Epidemiol 2014; 42: 224–233. © 2013 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Abstract – Objectives: To determine the pattern of fluid consumption, fluoride intake from the fluids and urinary fluoride excretion by children aged 1– 9 years in Kuwait, a nonfluoridated community. Methods: Using the cluster sampling technique, children aged 1–9 years were chosen from 2000 randomly selected households in Kuwait. Questionnaires were then administered to their mothers to determine the children’s daily fluid intake. Fluoride concentrations in tap water as well as all brands of bottled water and beverages consumed by the children were measured, using the fluoride ion-specific electrode. Fluoride excretion was determined in 400 randomly selected children, based on fluoride/creatinine ratio. Results: The mean daily fluid consumption by the children was high, being 1115–1545 ml. About 40% of the fluid intake was plain (tap and bottled) water and approximately 10% of the children drank bottled water exclusively. Fluoride concentration in tap water was low (0.04 SD 0.02 ppm), but was higher in bottled water (0.28 SD 0.40 ppm). Mean daily fluoride ingestion from fluids was 0.013–0.018 mg/kg body weight (bw). Even after allowing for fluoride ingestion from other sources, mean daily fluoride ingestion was still below 0.1 mg/kg bw set by the United States of America Institute of Medicine as the lowest-observed-adverse-effect level for moderate enamel fluorosis in children aged up to 8 years. Furthermore, the mean daily urinary fluoride excretion of 128–220 lg was below the provisional standard of 360–480 lg for optimal fluoride usage by children aged 3–5 years. Conclusion: Fluoride ingestion from fluids and urinary fluoride excretion by the children were below the recommendations for optimal fluoride usage. Thus, there is room for an upward adjustment of fluoride level in public drinking water supplies in Kuwait, as a caries preventive measure.
Exposure to fluoride from drinking water is a proven public health measure against dental caries (1–4). However, excessive fluoride ingestion during the period of tooth mineralization causes dental fluorosis (5–7). Appropriate fluoride concentration in drinking water varies in different parts of the world, depending on the climatic conditions. Galagan and Vermillion (8) suggested a formula for calculating appropriate fluoride level in drinking doi: 10.1111/cdoe.12081
Key words: cariology; epidemiology; fluoride; prevention; public health Enosakhare S. Akpata, Faculty of Dentistry, Kuwait University, P.O. Box 24923, Safat 13110, Kuwait Tel.: 965 2 498 6741 Fax: 965 2 498 6741 e-mail: [email protected] Submitted 29 November 2012; accepted 21 September 2013
water, based on ambient temperature. Thus, the appropriate fluoride concentration of approximately 0.7 ppm has been suggested for countries with a mean maximum annual ambient temperature above 27°C. Nevertheless, severe dental fluorosis has been observed in some tropical communities exposed to apparently appropriate fluoride level in drinking water. For example, in Saudi Arabia, severe dental
224
Fluoride ingestion and excretion in Kuwait
fluorosis was observed in some children exposed to well drinking water containing 0.5–0.8 ppm (9). Similar observations were reported in Senegal (10) and Sudan (11). It has been suggested that the high vulnerability to dental fluorosis in these tropical countries may be due to increased water consumption, and consequently excessive fluoride ingestion. Besides, Sohn et al. (12) recently showed that fluoride ingestion is related to fluid consumption, the patterns of which is undergoing changes in different parts of the world (13, 14). Other sources of fluoride ingestion include sea foods (15), toothpastes (16), infant formulas, and fluoride supplements (17, 18). Caries experience has been reported to be high among children in Kuwait and may be on the increase. A review of cross-sectional studies carried out on the occurrence of dental caries in the country between 1982 and 1994 showed a trend toward an increase in caries experience in the primary and permanent dentitions (19). Moreover, Vigild et al. (20) reported that caries had increased among children in Kuwait between 1982 and 1993. In a more recent national survey, Al-Mutawa et al. (21) reported caries prevalence to be 81.4% and 75.6% among children aged 5 and 6 years, respectively, and corresponding decayed and filled primary teeth (dft)/decayed and filled surfaces (dfs) to be 4.6/9.7 and 4.6/9.9, respectively. In contrast, caries prevalence among ethnic Danish children was reported to be 53% at age 7 years, while mean dmft was 3.5 (22). Fluoridation of public water supplies was discontinued in Kuwait in 1980; in view of the high caries experience among children in the country, an upward adjustment of the fluoride level in drinking water might be a viable caries preventive measure. If this were to be carried out, knowledge of fluoride ingestion from water and other fluids, as well as urinary fluoride excretion would be desirable, especially as these parameters appear to be unknown in Arab children during the period of tooth mineralization. The aim of this work, therefore, was to determine the pattern of fluid consumption, fluoride intake from ingested fluids, and daily urinary fluoride excretion by children aged 1–9 years in Kuwait, a nonfluoridated Arab community.
Subjects and methods Subjects Using the cluster sampling technique, children aged 1–9 years were chosen from about 2000
households from all the six governorates in Kuwait. Each of the governorates in the State of Kuwait is divided into suburbs or localities, and each suburb into blocks. Each block contains a certain number of streets. From each governorate, three suburbs were randomly selected, and from each suburb, three streets randomly chosen. Alternate houses on each selected street were then visited. The number of children included in the sample was calculated proportionally from the population of each governorate. The total number of children sampled from Ahmadi, Capital, Farwaniya, Hawalli, Jahra, and Mubarak Al-Kabeer governorates were 333, 153, 270, 756, 243, and 261, respectively, making a total of 2016 children sampled during each season (winter and summer). However, the mean number of mothers interviewed during each of the two seasons was 2022. The sample size was based on previous studies (13, 23, 24) to ensure a and b errors of 5% in the measurement of daily fluid ingestion (0.1 l), proportion (5%) of fluid intake that was plain water and fluoride ingestion (0.001 mg). Only one child was selected from each household. The children were in good general health and were not on any medication or fluoride supplements. If there was more than one child of the desired age in the selected household, one of them was chosen by ballot. If the mother in the household refused to participate or there was no child of the desired age, the interviewer proceeded to the next household. The refusal rate was approximately 20%. Approval for the research was obtained from the Ethics Committee of the Health Sciences Center, Kuwait University. The purpose of the research was explained to the parents of the selected children, and they gave their consent by signing the consent form. Demographic information was obtained from the children’s parents.
Daily fluid consumption and oral hygiene practices Prior to the study, four fieldworkers, who had previous experience with household surveys, were trained in the use of the questionnaire designed for the present study. Each of them then visited five households in different governorates in Kuwait. The research team discussed with them their findings and resolved potential problems. The pretested questionnaire was administered to mothers of the selected children by the trained interviewers. The questionnaire inquired about the types and quantities of all fluids ingested by the
225
Akpata et al.
children within the previous 24 hours, that is, the fluid intake before breakfast, at breakfast, between breakfast and lunch, at lunch, between lunch and dinner, at dinner and after dinner before going to bed. Household measures and color photographs of various sizes of bottled water and beverages were used to estimate volumes. Administration of the questionnaire was for 3 days in summer and another 3 days in winter. This was repeated for the same child after an interval of at least 3 days during each of the seasons. The results for the first and second interviews were averaged for each of the seasons. The questionnaire also inquired about the children’s oral hygiene practices, including the type, amount, and frequency of fluoride toothpaste use. The size of toothpaste was measured in terms of toothbrush lengths, illustrated by color photographs.
Measurement of weights and heights The children’s weights were measured with calibrated bathroom scales (Terraillon, Croissy-SurSeine Cedex, France) to the nearest 0.5 kilogram. While measuring their weights, the children put on light clothing and were barefooted. With the children barefooted, and standing against a vertical wall, their heights were measured with a tape rule to the nearest 1 cm, with the child’s heels, back, and occiput touching a vertical wall, and a horizontal bar lowered to touch the vertex. From the measurements, body mass index (BMI) was calculated. BMI = weight (kg)/height (m2).
Fluoride concentrations in drinking water and beverages An experienced technician from the Faculty of Science, Kuwait University, carried out fluoride analysis of the samples. Preliminary analyses were carried out on 15 samples of plain water, beverages, and urine. Duplicates of the samples, to which the technician was blinded, were re-analyzed for fluoride two days later. Correlation plot showed very high correlation (r = 0.9, P < 0.0001) between the results of the two measurements, indicating high reproducibility. Tap water samples collected from households, different commercial brands of bottled water, cold, and hot beverages (including tea) consumed by the children were analyzed in duplicate for fluoride content, using the Fluoride Ion Selective Electrode, ISE (Model 960900; Thermo Electron Corporation,
226
MA, USA) in conjunction with an ISE Meter (Orion Model 720A; Thermo Electron Corporation, Beverly, MA, USA). Twenty-five milliliters of each sample was mixed with equal volume of Total Ionic Strength Adjusting Buffer II (TISAB II), before analysis. Fluoride standards ranging from 0.01 to 1 mg/l were used to calibrate the measurements. The tap water (about 200 ml) was obtained in polythene bottles from four taps within households in each of the six governorates in Kuwait. The polythene bottles were rinsed three times with the tap water before sample collection. Samples of bottled water and beverages were from the various brands bought from supermarkets and grocery stores in Kuwait. Tea infusions for the analysis were prepared by adding one tea bag to 100 ml of boiled de-ionized water, and the tea brewed for 2 minutes. Powdered beverages were reconstituted according to the manufacturers’ instructions/recommendation, using de-ionized water. Fluoride ingestion from tap water was calculated by multiplying fluoride concentration by the volume of water consumed. In the case of other fluids, fluoride ingestion was calculated for each child by multiplying the volume of the fluid consumed by the fluoride concentration in that particular brand of fluid.
Estimation of urinary fluoride excretion Early morning spot urine sample was obtained from the child at every fifth household visited. The fieldworker gave the mother 60-ml polythene bottle at the first visit to the household and collected the urine sample at the second visit. Because of the practical difficulty in collecting 24-hour urine samples from many children in a study of this nature, we estimated 24-hour urinary fluoride excretion from the fluoride/creatinine ratio of the early morning spot urine samples. The samples were stored at 20°C, until analyzed in triplicates, using the ISE after adding TISAB II. Urinary creatinine concentrations of the spot urine samples were measured by the kinetic Jaffe reaction (25), using the Beckman Synchron LX20 automated analyzer (Beckman Corporation, Brea, CA, USA). From the concentrations of urinary fluoride and urinary creatinine measurements, the fluoride/creatinine ratio was calculated. The subject’s 24-hour urinary fluoride excretion was then calculated by multiplying the ratio by the standard creatinine value [15 mg/l/kg body weight (bw)/ day] and the child’s bw (26).
Fluoride ingestion and excretion in Kuwait
Statistical analysis
the mothers interviewed was rather high: about 41% had tertiary/university education, while only about 10% had primary or no formal education. Over 98% earned at least the recommended minimum salary (about $1500 per month) for families in Kuwait. Only about 3% of the children were obese (BMI ≥ 25). The mean maximum annual temperature during the summer months was 44°C and in winter 22°C.
Descriptive statistics were presented as frequency, percentages, means, and standard deviations. ANOVA was used to determine the statistical significance of the differences between fluid consumption, fluoride ingestion, and excretion among children of different sexes, age groups, nationality, and BMI; Multiple logistic regression analysis was used to determine factors associated with high fluoride ingestion (cutoff point set at 0.02 mg/kg bw/day, i.e. the 75th percentile of the children’s fluoride ingestion). The data management and statistical analysis were carried out using the statistical software SPSS-PC version 18.0 (IBM, New York, NY, USA).
Fluid consumption Mean daily fluid consumption among children in Kuwait was relatively high (Table 1), being 1129 302, 1372 321, and 1599 311 ml in summer and 1149 231, 1331 271, and 1538 282 ml in winter at ages 1–3, 4–6, and 7–9 years, respectively. ANOVA showed the age and seasonal differences to be statistically significant (P < 0.05). The only exceptions were the consumption of bottled water in which there were no significant age differences, and carbonated drinks in which there were no seasonal differences. In general, standard deviations were rather high, indicating the great variability in fluid consumption. Plain water constituted 41% of fluid intake by 1–9-year-old children in Kuwait in summer, of which about 6% was bottled water, the remainder being tap water. In winter, however, plain water constituted 36% of fluid intake, of which about 5% was bottled water (Table 1). About 11% of the children drank bottled water exclusively, while 2% never drank it. Most of the children (87%) drank bottled water occasionally.
Results Demographic characteristics The questionnaire study was carried out in 2009– 2011, and 2018 mothers of children aged 1–9 years were interviewed in summer and 2026 in winter. The ratio of boys to girls was 52:48, and their distribution in the age groups 1–3, 4–6, and 7–9 years was reasonably even. About 95% of the children were Arabs, and about 60% Kuwaitis (Kuwaitis are Arabs and non-Kuwaiti Arabs include those from other countries in the Middle East, such as Saudi Arabia, Egypt, United Arab Emirates and Qatar). Among non-Arabs in Kuwait are Indians, Pakistanis and Westerners). The mean bws of the children aged 1–3, 4–6, and 7–9 years were 13.2, 20.7, and 30.6 kg, respectively. The level of education of
Table 1. Frequency distribution of daily fluid consumption (ml) among children in Kuwait, by age groups and season*, based on questionnaire interviews Summer
Winter
Age groups (years)
Age groups (years)
1–3
4–6
7–9
1–3
4–6
7–9
N = 561
N = 704
N = 753
N = 613
N = 726
N = 687
Fluids
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Mean
SD
Total fluid Plain water Tap water Bottled water Carbonated beverages Fruit juices and milk Hot beverage (tea) Hot beverages (other than tea)
1129 430 335 95 43 450 10 195
302 186 239 169 62 213 34 215
1372 584 515 69 116 543 58 69
321 226 264 178 94 167 79 84
1599 672 586 86 183 572 118 53
311 238 293 211 119 186 102 82
1149 367 300 66 48 410 16 309
231 135 187 137 64 211 42 249
1331 494 422 71 120 510 80 129
271 171 236 166 98 164 92 145
1538 581 512 69 169 527 155 106
282 198 264 181 111 163 109 127
*Two-way ANOVA (age groups and seasons), P < 0.05, except for carbonated drinks (P = 0.475), and bottled water (P = 0.278).
227
Akpata et al.
Children whose mothers had higher formal educational background and income consumed significantly less fluid (P < 0.05). However, there was no significant difference in fluid consumption between the Kuwaitis and non-Kuwaitis, irrespective of the type of fluid. There was a significant association between beverage ingestion and obesity although the correlation was weak (r = 0.16, P < 0.001).
Fluoride concentration in fluids The children consumed many brands of bottled water and beverages: 15 brands of bottled water, 30 of juices, 26 soft drinks of which eight were carbonated, eight of hot beverages including tea, and 15 of milk products, most of which were simply milk flavored, making a total of 94 brands of bottled water and beverages. Table 2 summarizes the fluoride levels in bottled water, liquid, and powdered beverages. It can be seen that over 66% of the bottled water and beverages had fluoride concentrations of 0.01–0.099 ppm. None of the brands of bottled water had fluoride level above 1.10 ppm of fluoride, while
