Arsenic-Contaminated Water and Extent of Acute Childhood Malnutrition (Wasting) in Rural Bangladesh

Transcription

1 Environmental Sciences, 12, 5 (2005) K. Minamoto et al. 283 MYU Tokyo ES599 Arsenic-Contaminated Water and Extent of Acute Childhood Malnutrition (Wasting) in Rural Bangladesh Keiko Minamoto *, C.G. Nicholas Mascie-Taylor 1, Kazuhiko Moji 2, Enamul Karim 3 and Mahmudur Rahman 4 Department of Preventive and Environmental Medicine, Graduate School of Medical and Pharmaceutical Sciences, Kumamoto University, Honjo, Kumamoto , Japan 1 Department of Biological Anthropology, University of Cambridge, Cambridge, UK 2 Institute of Tropical Medicine, Nagasaki University, Nagasaki , Japan 3 Health Life Sciences Partnership, Dhaka, Bangladesh 4 Institute of Epidemiology, Disease Control and Research, Dhaka, Bangladesh (Received April 27, 2005; accepted September 6, 2005) Key words: arsenic contamination, child s nutritional status, acute malnutrition, intestinal parasites To investigate whether children s nutritional status, especially the extent of wasting [acute malnutrition as measured by weight-for-height Z (WHZ) score] is associated with arsenic contamination of drinking water, a survey was carried out in two discrete rural areas located less than 30 km southeast and northeast of Dhaka, the capital of Bangladesh. 761 children between 7 and 14 years of age were randomly selected, one child per household. The arsenic concentration in the tube well water used by each child and the prevalence and intensity of infection of three geo-helminths [Ascaris lumbricoides, Trichuris trichiura and hookworm (Ancylostoma duodenale or Necator americanus)] were determined. Each child had their height and weight measured and WHZ score determined using the World Health Organization (WHO) reference values. Basic socio-demographic data and knowledge of arsenic poisoning were collected using a questionnaire. 62.2% of the households studied were using well water with arsenic concentrations above the 0.01 mg/l WHO guideline (i.e. arsenic-contaminated households). The mean WHZ score of all the 761 children was (sd 1.227) but children living in contaminated households were significantly more wasted than children living in noncontaminated households (mean difference = 0.361, p < 0.001). When the effects of the differences in socio-economic status and prevalence of geohelminths were also taken into account, the difference in means remained highly significant ( 0.330, p = 0.006). This study suggests that arsenic contamination has a negative impact on children s acute nutritional status. * may20@kaiju.medic.kumamoto-u.ac.jp 283

2 284 Environmental Sciences, 12, 5 (2005) K. Minamoto et al. 1. Introduction Groundwater contamination by naturally occurring arsenic was first found in Argentina at the beginning of the last century. Since then, arsenic-contaminated groundwater has been reported in many countries. (1) In Bangladesh, the first arsenic-contaminated well water was discovered in the Nawabganj district near the border with India in 1993 (2,3) and since then contaminated wells have been found in nearly all districts. Groundwater is free of pathogenic microorganisms and the provision of tube wells is a cost-effective means of providing clean water supplies to the scattered rural Bangladeshi population. The number of tube wells installed privately has grown since the 1970s and reached over 10 million. (4) However, recent national surveys have found that two-thirds of all tube wells are above the 0.01 mg/l World Health Organization (WHO) cut-off arsenic contamination. (4) Chronic arsenic poisoning effects on adults have been well documented and it is well known that it increases the risk of developing various skin disorders as primary symptoms (5,6) and many types of disease such as hypertension, peripheral vascular disease, cardiovascular disease and diabetes mellitus, while continued arsenic poisoning can result in lung, bladder, kidney and skin cancers. (1) However, the effect of arsenic contamination on children has not been widely reported and in addition very little is known about the relationship between arsenic poisoning, children s growth and nutritional status. One of the few studies conducted was a follow-up of infants exposed to arsenic-contaminated milk in Japan. The infants, the victims of human error, were exposed to mg per day, which caused subacute poisoning and 130 deaths. (7) The survivors, who were examined about 15 years after the exposure, were, on average, significantly shorter than the nonexposed children. A small recent study carried out in Bangladesh provided evidence that children chronically exposed to arsenic are more stunted in growth than nonexposed children. (8) In order to examine the relationship between the extent of acute malnutrition [wasting, weight-for-height Z (WHZ) score] and arsenic poisoning at the preclinical level, a study was conducted in two rural areas of Bangladesh. Because nutritional status is also associated with differences in the prevalence and intensity of geo-helminth infection and the socioeconomic status between families, these variables were also taken into account in this study. 2. Subjects and Methods A cross-sectional study in two discrete geographical sites (Sonargaon, which is located 30 km southeast of Dhaka, the capital of Bangladesh, and Palash, located 25 km northeast of the capital) (Fig. 1) was carried out in August October, Ethical clearance was obtained from the Bangladesh Medical Research Council. 761 children between 7 and 14 years of age were randomly selected on a household basis to participate in the study. The arsenic concentration in the tube well water used by each child was determined using a field kit designed by the National Institute of Preventive and Social Medicine. (9,10) Each child provided a faecal sample so that the prevalences and intensities of infection (eggs/g in faecal samples, using the Kato-Katz method) of the three geo-helminths [Ascaris lumbricoides, Trichuris trichiura and hookworm (Ancylostoma duodenale or Necator americanus)] could

3 Environmental Sciences, 12, 5 (2005) K. Minamoto et al. 285 Fig. 1. Locations of Sonargaon and Palash. be determined. Each child had their height and weight measured and the extent of wasting (WHZ score) was determined using the growth reference values of WHO. (11) Basic sociodemographic data of the households studied were collected and one of the parents of the index child was questioned regarding awareness of arsenic problems and knowledge of skin symptoms caused by arsenic poisoning. The index child and the parent (mainly the mother) were examined for arsenic poisoning symptoms in their extremities and information on skin symptoms of close family members was also collected. Statistical analyses were conducted using SPSS version 10.0.

4 286 Environmental Sciences, 12, 5 (2005) K. Minamoto et al. 3. Results 3.1 Arsenic contamination, socio-economic status and arsenic symptoms among members of index child s family Of the 761 children, 448 lived in Sonargaon and 313 in Palash. 375 of the children (49.3%) were male. There was a significantly higher number of male children in Palash than in Sonargaon (54.0% versus 46.0%, p = 0.03). Children from Palash were significantly older than those from Sonargaon [mean (sd 19.0) and (sd 19.1) months respectively, p < 0.001]. 62.2% of the children were using well water with arsenic concentrations above the 0.01 mg/l WHO standard. However, contaminated households were primarily located in Sonargaon where nearly all households (96.7%) were using water with arsenic concentrations above the WHO standard compared with only 12.8% of households in Palash (p < 0.001). In addition, the mean arsenic concentration was significantly higher in Sonargaon than in Palash (p < 0.001) (Table1). Fathers in Palash had higher education levels but were more likely to be employed as daily labourers than fathers in Sonargaon. The ownerships of television and radio were very similar in both sites (18.3% and 26.7%, respectively). Households in Palash were more likely to have their own tube well and were less likely to have a latrine than those in Sonargaon. Children in Palash were more likely to have Ascaris and hookworm infections than children from Sonargaon. However, no significant difference was found in Trichuris prevalence (Table 2). Neither the index child nor their parents were found with any skin symptoms characteristic of arsenic poisoning and there were no reports of such skin symptoms among their close family members. Table 1 Comparison of age and sex of index child and arsenic concentration in the tube well water used by the index child by site. Sonargaon Palash Total p value (N = 448) (N = 313) (N = 761) Mean of age (months) < (SD) (19.0) (19.1) (21.1) Number of male children (%) (46.0%) (54.0%) (49.3%) Number of children using contaminated tube well water above 0.01mg/L WHO standard < (%) (96.7%) (12.8%) (62.2%) Arsenic concentration (mg/l) of the tube well water used by each index child Mean < (SD) (0.158) (0.027) (0.149) Median Range

5 Environmental Sciences, 12, 5 (2005) K. Minamoto et al. 287 Table 2 Comparison of socio-economic status and prevalence of intestinal parasites by site. Sonargaon Palash Total N % N % N % p value Father s education < None years years Secondary Higher & graduate Total Father s occupation < Daily labourer Agricultural worker Non-agricultural worker Regular service worker Businessman Total Owns a television No Yes Total Owns a radio n.s. No Yes Total Owns a tube well < No Yes Total Latrine type < None Hanging or pit Slab Western style Total Child positive for Ascaris eggs *1 < Yes No Total Child positive for Trichuris eggs *2 n.s. Yes No Total Child positive for hookworm eggs *3 < Yes No Total *1 Ascaris lumbricoides *2 Trichuris trichiura *3 Ancylostoma duodenale or Necator americanus n.s.

6 288 Environmental Sciences, 12, 5 (2005) K. Minamoto et al. 3.2 Knowledge of arsenic poisoning Of the 761 parents questioned, just under a third (30.3%) said that they had heard about arsenic poisoning. There was a significant difference in such knowledge between the two areas (Sonargaon 33.1% versus Palash 26.3%, p = 0.044) but not between contaminated and noncontaminated households. 107 parents (14.1%) correctly knew about skin symptoms resulting from arsenic poisoning (e.g., blackish spot, hardening of the skin, ulceration, itching and reddish skin) and there was significantly greater knowledge of the symptoms among parents living in Sonargaon than in Palash (16.5% versus 10.5%, p = 0.020). 3.3 Nutritional status and arsenic contamination The vast majority of children did not have birth certificates and age was estimated using local calendars. Thus, there were bound to be inaccuracies as well as rounding (to the nearest 6 months or whole year). Since height-for-age and weight-for-age Z scores rely on accurate ages, the analyses described here focussed only on weight-for-height, i.e., a measure of acute malnutrition. The mean WHZ score of all the 761 children was (sd 1.227) indicative of significant levels of malnutrition and 12.5% of children had a Z score < 2.0 which is the cut-off for severe malnutrition. Since WHZ scores varied by sex and age, analysis were undertaken which adjusted for these variables before testing for site differences. The results presented in Table 3 show that the mean WHZ score was significantly lower in Sonargaon than in Palash (mean difference 0.283, p = 0.004) and in contaminated households than in noncontaminated households (mean difference 0.361, p < 0.001). The sequential analysis of variance was used to remove the effects of age and sex, the background socio-economic variables and geo-helminth infection status (prevalence), and the difference between contaminated and noncontaminated households remained significant (mean difference 0.330, p = 0.006). The use of the intensities of infection, instead of prevalence, gave essentially similar results (mean difference 0.327, p = 0.007). Table 3 Comparison of mean WHZ (weight-for-height Z) score of index child by site and between contaminated and noncontaminated households divided by 0.01 mg/l WHO standard. Corrected for age and sex *1 Corrected for age, sex, socio-economic status and geo-helminth infection status *1 (N = 737) (N = 600) Mean Difference in mean *2 p value Difference in mean *2 p value Site n.s. Sonargaon Palash WHO standard < > 0.01 mg/l mg/l *1 Corrected using linear regression analysis. *2 The reference value is 0.

7 Environmental Sciences, 12, 5 (2005) K. Minamoto et al. 289 In addition, the effect of three different categories of arsenic contamination on mean WHZ score was examined (Fig. 2) and increasing contamination was significantly associated (p < 0.001) with decreasing mean WHZ score after correcting for age and sex (model 1 in Fig. 2). After correcting for age, sex, the background socio-economic variables and infection status by the three kinds of intestinal geo-helminth (model 2 in Fig. 2), the mean differences in WHZ by arsenic contamination were still significant (p = 0.021) although the main difference was between and higher levels of contamination. 4. Discussion Nearly 800 million of the world s population are chronically malnourished, of whom 200 million are moderately to severely undernourished and 70 million are severely malnourished. (12) The causes of malnutrition are complex and include nutritional deficiencies in vitamin A, calcium, iron and iodine. However, as is increasingly being recognised, environmental lead, arsenic, mercury and other metals that enter the food chain can seriously deplete body stores of iron, vitamin C and essential nutrients leading to decreased immune defences, intrauterine growth retardation, impaired psycho-social faculties and other disabilities associated with malnutrition. Arsenic produces various stress responses in mammalian cells, including metabolic abnormalities accompanied by growth inhibition and eventually apoptosis. (13) Cells exposed to arsenic show morphological alterations with the disruption of cytoskeletal structures responsible for cellular integrity, shape and locomotion. *1 The mean WHZ score of the index child using tube well water with an arsenic concentration of mg/l is set to 0. *2 Model 1 is corrected for age and sex and Model 2 is corrected for age, sex, socio-economic status and geo-helminth infection status using linear regression analysis. Fig. 2. Mean WHZ (weight-for-height) scores of index child for three categories of increasing arsenic concentration of tube well water used by each index child.

8 290 Environmental Sciences, 12, 5 (2005) K. Minamoto et al. There is evidence that arsenic affects the outcome of pregnancy with higher rates of spontaneous abortion and stillbirth in exposed mothers as well as lower birth weight. (14 16) However, the effects of arsenic on children s growth and nutrition have been rarely reported. A follow-up study of Japanese infants exposed to arsenic ( mg/day) through intake of contaminated powder milk showed that as teenagers they were, on average, shorter and of lower weight. (7) However, these infants were only exposed to subacute arsenic poisoning and not to chronic exposure as found in Bangladesh. Depending on the extent and degree of arsenic stress, it might be expected that longerterm growth and development would be impaired. The present finding of a lower mean WHZ score among children living in contaminated, compared with noncontaminated households, as well as the trend for a lower mean WHZ score with increasing arsenic concentration, is also suggestive of a longer-term negative impact of arsenic stress. A small study conducted in Bangladesh revealed a lower mean weight-for-age Z score among children exposed to arsenic-contaminated well water. (8) Recently, Watanabe et al. (17) have reported marginally significant differences in median WHZ score between the lowestand the highest-exposure quartiles in Bangladeshi girls, although the sample size was very small. These two studies, as well as the present one, being cross-sectional, cannot provide proof of growth retardation in children exposed to elevated levels of arsenic; thus, longitudinal studies that take into account differences in nutritional intake and disease burdens, particularly from geo-helminths, are required. Finally, there is evidence that parents living in the contaminated area were more aware of the problems of arsenic exposure than parents living in the noncontaminated area. However even in the contaminated area (Sonargaon) only about 30% of the parents of the index child knew about arsenic problems. One of the reasons is that there were no physical symptoms of arsenic poisoning, i.e., only preclinical. If there is a safe tube well close to each child s residence, it is recommended not to use contaminated water. This is the first action to mitigate the arsenic problem. However, in Sonargaon, safe water with arsenic concentration below the 0.01 mg/l WHO standard was available to only a very small percentage (3.3%) of households and although not all tube wells in Sonargaon were examined, it is unlikely that safe tube well water can be obtained by every household. In conclusion, this study provided evidence that arsenic stress at the preclinical level can affect child s growth and nutrition. Even in the contaminated area, knowledge of the arsenic problem was limited and, apparently, it is difficult for every resident to obtain safe water. Until such time that safe water is available, and the Government of Bangladesh is actively considering a number of intervention measures, the nationwide extent of the contamination problem will remain critical. Acknowledgments We thank all the children and their families for agreeing to participate in this study. All children received anthelmintic treatment at the end of the study if found positive for any of the geo-helminths. Sincere thanks to Dr. Makoto Futatsuka (Professor emeritus, Kumamoto University), President of Kyushu University of Nursing and Social Welfare, Tamana, Kumamoto, Japan, for his very helpful advice and encouragement.

9 Environmental Sciences, 12, 5 (2005) K. Minamoto et al. 291 References 1 IPCS (2001): Effects on humans. In: Environmental Health Criteria 224 Arsenic and Arsenic Compounds, 2nd Ed. WHO, Geneva, pp Khan, A.W., Ahmad, S.A. and Sayed, M.H.S.U. (1997): Arsenic contamination in groundwater and its effect on human health with particular reference to Bangladesh. Journal of Preventive and Social Medicine 16: Smith, A.H., Lingas, E.O. and Rahman, M. (2000): Contamination of drinking-water by arsenic in Bangladesh: a public health emergency. Bull. of the WHO 78: British Geological Survey and Department of Public Health Engineering (2001): Arsenic contamination of groundwater in Bangladesh. In: D.G. Kinniburgh and P.L. Smedley Eds.: BGS Technical Report WC/00/19. British Geological Survey, Keyworth, England, Vol. 2 (Final Report). 5 Schwartz, R.A. (1997): Arsenic and the skin. International Journal of Dermatology 36: Ahmad, S.A., Sayed, S.U.M.H., Hadi, S.A., Faruquee, M.H., Khan, M.H., Jalil, A., Ahmad, R.M. and Khan, A.W. (1999): Arsenicosis in a village in Bangladesh. International Journal of Environmental Health Research 9: Yamashita, N., Doi, M. and Nishio, M. (1972): Recent observation of Kyoto children poisoned by arsenic tainted Morinaga Dry Milk. Japanese Journal of Hygiene (Nihon Eiseigaku Zasshi) 27: (in Japanese). 8 Ahmad, T.S. (2002): Effects of chronic arsenic exposure on nutritional status of children (6 12 yrs.) in a selected area of Bangladesh. In: Research Studies on Health Impact of Arsenic exposure. Bangladesh Medical Research Council (BMRC), Dhaka, pp Ahmad, S.H., Sayed, S.U.M.H., Hadi, S.A. and Khan, A.W. (1997): Modified arsenic field test kit: a cheap and easy device for detection of arsenic in water. Journal of Preventive & Social Medicine 16: Pande, S.P., Deshpande, L.S. and Kaul, S.N. (2001): Laboratory and field assessment of arsenic testing filed kits in Bangladesh and West Bengal. Environmental Monitoring and Assessment 68: WHO (1983): Measuring Change in Nutritional Status. WHO, Geneva. 12 Iyengar, G.V. and Nair, P.P. (2000): Global outlook on nutrition and the environment: meeting the challenges of the next millennium. Science of the Total Environment 249: Bernstam, L. and Nriagu, J. (2000): Molecular aspects of arsenic stress. Journal of Toxicology & Environmental Health Part B: Critical Reviews 3: Nordstrom, S., Beckman, L. and Nordenson, I. (1979): Occupational and environmental risks in and around a smelter in northern Sweden V: spontaneous abortion among female employees and decreased birth weight in their offspring. Hereditas 90: Tabacova, S., Baird, D.D., Balabaeva, L., Lolova, D. and Petrova, I. (1994): Placental arsenic and cadmium in relation to lipid peroxides and glutathione levels in maternal-infant pairs from a copper smelter area. Placenta 15: Ahmad, S.K., Sayed, S.U.M.H., Khan, S.B.M.H., Faruquee, M.H., Jalil, A., Hadi, S.A. and Talukder, H.K. (2001): Arsenic in drinking water and pregnancy outcomes. Environmental Health Perspectives 109: Watanabe, C., Inaoka, T., Matusi, T., Ishigaki, K., Murayama, N. and Ohtsuka, R. (2003): Effects of arsenic on younger generation. Journal of Environmental Science and Health. Part A-Toxic/ Hazardous Substances & Environmental Engineering A 38: