Background
Heavy metals are ubiquitous in the environment, and the greatest burden of toxic exposure is often borne by children in low-income countries, where biomonitoring data are scant and little regulation exists for waste [
1]. Many of these metals are known to be neurotoxic to developing brains, but lack of knowledge and expensive testing supplies has hindered measurement and assessment of their burden in children in low-income countries.
A growing body of evidence suggests that exposure to multiple different heavy metals and metal mixtures can have a profound neurotoxic effect [
2‐
4]. Lead exposure is known to cause intellectual deficit, neurodevelopmental delay, and increased risk for cognitive disorders including ADHD [
5]. Manganese excess causes direct damage to neuronal cells [
6], while arsenic toxicity in drinking water was shown to dose dependently inhibit antioxidant activity and increase oxidative stress in the rat brain [
7]. Even excess of nutritional metals such as iron and copper can cause neuronal damage [
8]. Although these effects have been largely unstudied in children in low-income countries, it is in these regions where heavy metal exposure may have its most damaging neurobehavioral consequences, both independently and also via interaction with nutritional deficiency and infectious illness. While these relationships are complex, the high risk for heavy metal exposure for children in low-income countries and the known neurotoxic effects of heavy metal exposures compel monitoring of blood metal levels and examination of possible environmental determinants.
To assess the burden of heavy metal exposure in healthy Ugandan children, we recently conducted a community-based survey of blood levels of 13 heavy metals in 100 children younger than five years of age living in the Katanga urban settlement of Kampala. We also administered an environmental questionnaire to identify factors associated with metal exposure. The overall aim of the study was to determine the burden of heavy metal exposure to children in this community and to identify possible sources of exposure. A finding of prevalent metal exposure would highlight a potentially remediable cause of morbidity in this population and lay the foundation for future longitudinal studies with neurobehavioral outcomes.
Results
Of the 100 children enrolled, there were slightly more males than females, and the average age was just over two years (Table
2, Additional files
1 and
2). Nearly one in four children was stunted (height-for-age Z-score < − 2). Three percent of children were wasted (weight-for-height Z-score < − 2). One child was diagnosed with and treated for
P. falciparum malaria. Most mothers had completed upper primary school. One-third of households had a radio, 40% had a television, and more than half had electricity.
Table 2
Sample characteristicsa
n |
Age, monthsb | 27.9 (15.1) |
Sex, n (% M) | 53 (53.0) |
Height-for-age Z-scorec | −0.76 (1.8) |
Stunted, n (%)c | 23 (23.5) |
Weight-for-height Zd | −0.33 (1.0) |
Wasted, n (%)d | 3 (3.1%) |
Weight-for-age Z-scored | −0.73 (1.2) |
Underweight, n (%) | 15 (15) |
Malaria-positive, n (%) | 1 (1.0) |
Highest attained education of mother or primary caregivere |
Never went to school | 9 (9.1) |
Lower primary school | 14 (14.1) |
Upper primary school | 42 (42.4) |
O′ level secondary school | 26 (26.3) |
A’ level secondary school | 3 (3.0) |
Tertiary level | 4 (4.0) |
Household possessionse |
Radio, n (%) | 33 (33.3) |
TV, n (%) | 41 (41.4) |
Electricity, n (%) | 57 (57.6) |
Heavy metal exposure was prevalent. Of the 13 metals measured, blood levels of six were elevated in this community-based sample (Table
3). All enrolled children had elevated blood antimony levels. All but three children had a blood lead level of greater than 2 μg/dL, and approximately one in three had an elevated blood manganese level. Nearly 20% had an elevated concentration of cadmium and cobalt, and more one in ten children had elevated blood copper.
Table 3
Blood concentrations of heavy metals in Ugandan children and prevalence of elevated levels
Antimony, μg/L | 100 | 5.71 | 6.61 | 7.29 | 8.31 | 8.76 | 99 (99.0) |
Arsenic, μg/L | 100 | 0.15 | 0.22 | 0.33 | 0.47 | 0.58 | 0 (0.0) |
Barium, μg/L | 100 | 1.00 | 1.44 | 2.42 | 6.27 | 8.29 | 0 (0.0) |
Cadmium, μg/L | 100 | 0.04 | 0.09 | 0.13 | 0.17 | 0.23 | 17 (17.0) |
Cesium, μg/L | 100 | 0.46 | 0.57 | 0.77 | 0.96 | 1.04 | 0 (0.0) |
Chromium, μg/L | 99 | 0.50 | 0.59 | 0.81 | 1.30 | 1.98 | 7 (7.1) |
Cobalt, μg/L | 99 | 0.28 | 0.42 | 0.55 | 0.87 | 1.18 | 19 (19.2) |
Copper, μg/L | 100 | 1025.08 | 1165.51 | 1428.08 | 1579.53 | 1719.39 | 12 (12.0) |
Lead, μg/dL | 100 | 4.47 | 5.78 | 7.69 | 10.01 | 10.87 | 3 (3.0) [0 - < 2 μg/dL] 34 (34.0) [> = 2 - < 5 μg/dL] 53 (53.0) [> = 5 - < 10 μg/dL] 10 (10.0) [> = 10 μg/dL] |
Manganese, μg/L | 99 | 9.35 | 13.62 | 23.84 | 31.42 | 37.35 | 36 (36.4) |
Nickel, μg/L | 99 | 0.60 | 0.68 | 0.88 | 1.05 | 1.29 | 1 (1.0) |
Selenium, μg/L | 100 | 106.75 | 121.70 | 151.81 | 167.83 | 181.54 | 1 (1.0) |
Zinc, μg/L | 100 | 3019.95 | 3544.39 | 4284.47 | 4946.96 | 5512.64 | 6(6.0) |
The environmental questionnaire revealed some environmental factors associated with elevated blood metal concentrations (Table
4). Cement walls and floors were both associated with higher blood manganese. Children whose houses had windows tended to have higher blood antimony than children whose houses had no windows (
p = 0.08), while children whose drinking water came from a well or spring tended to have high blood lead (
p = 0.10) than children whose drinking water came from a community or home tap. Although only five children lived near a garbage landfill, these children had significantly higher blood cobalt than children who did not live near a landfill (
p = 0.01). Finally, the 16 children who attended school had significantly higher blood cadmium levels than children who did not attend school (
p < 0.01).
Table 4
Median whole blood heavy metal concentrations by selected environmental characteristics1
House |
Wall material3 |
Cement | 53 (55.2) | 6.7 (5.8, 7.3) | 0.75 | 0.09 (0.04, 0.14) | 0.78 | 0.39 (0.29, 0.53) | 0.84 | 1142 (997, 1429) | 0.62 | 5.6 (3.9, 6.9) | 0.22 | 17.2 (10.4, 27.1)a | 0.04 |
Mud/dung | 40 (43.8) | 6.4 (5.5, 7.5) | | 0.08 (0.04, 0.13) | | 0.46 (0.33, 0.62) | | 1160 (1061, 1437) | | 6.1 (4.7, 8.9) | | 12.2 (9.4, 19.0)b | |
Other | 3 (3.1) | 7.1 (6.6, 7.3) | | 0.11 (0.01, 0.14) | | 0.54 (0.21, 0.85) | | 1204 (945, 1260) | | 6.6 (3.7, 9.4) | | 9.0 (8.7, 20.8) a,b | |
Painted walls,3 doors, windows |
No | 67 (69.8) | 6.6 (5.7, 7.4) | 0.56 | 0.09 (0.04, 0.14) | 0.42 | 0.48 (0.31, 0.65) | 0.08 | 1208 (1072, 1446) | 0.84 | 6.1 (4.7, 8.8) | 0.50 | 13.2 (9.4, 21.6) | 0.87 |
Yes | 29 (30.2) | 6.6 (5.5, 7.1) | | 0.06 (0.03, 0.12) | | 0.36 (0.28, 0.45) | | 1107 (977, 1386) | | 4.5 (3.9, 5.8) | | 15.5 (9.3, 24.7) | |
Floor material |
Cement4 | 51 (54.3) | 6.6 (5.7, 7.2) | 0.51 | 0.09 (0.04, 0.14) | 0.78 | 0.40 (0.28, 0.52) | 0.53 | 1135 (987, 1463) | 0.39 | 5.6 (4.0, 6.3) | 0.89 | 15.8 (9.3, 27.6) | 0.04 |
Mud/ dung/clay | 43 (45.7) | 6.6 (5.7, 7.4) | | 0.08 (0.04, 0.13) | | 0.45 (0.32, 0.58) | | 1171 (1050, 1355) | | 6.1 (4.7, 8.9) | | 12.0 (9.0, 18.5) | |
Windows5 |
No | 52 (67.5) | 6.4 (5.5, 7.1) | 0.08 | 0.09 (0.04, 0.14) | 0.75 | 0.40 (0.29, 0.54) | 0.77 | 1114 (978, 1345) | 0.28 | 5.7 (4.1, 7.7) | 0.30 | 12.9 (8.6, 19.7) | 0.26 |
Yes | 25 (32.5) | 7.0 (6.5, 7.7) | | 0.08 (0.06, 0.12) | | 0.41 (0.31, 0.54) | | 1260 (1106, 1260) | | 6.1 (4.6, 7.2) | | 15.1 (10.5, 19.6) | |
Water/Cooking |
Drinking water |
Community tap | 66 (66.0) | 6.6 (5.8, 7.5) | 0.88 | 0.08 (0.04, 0.13) | 0.59 | 0.40 (0.28, 0.54) | 0.49 | 1133 (997, 1382) | 0.52 | 5.6 (3.9, 6.4) | 0.10 | 13.0 (9.3, 20.9) | 0.36 |
Tap in home | 18 (18.0) | 6.4 (5.7, 6.8) | | 0.09 (0.03, 0.15) | | 0.41 (0.27, 0.63) | | 1227 (1063, 1467) | | 5.9 (4.7, 8.7) | | 13.5 (10.5, 15.8) | |
Well or spring | 16 (16.0) | 6.6 (5.6, 7.3) | | 0.09 (0.05, 0.15) | | 0.48 (0.33, 0.55) | | 1492 (1092, 1493) | | 7.1 (5.7, 10.1) | | 20.2 (9.3, 30.1) | |
Cooking where?3 |
Inside | 53 (55.2) | 6.5 (5.7, 7.3) | 0.44 | 0.09 (0.04, 0.14) | 0.40 | 0.45 (0.31, 0.55) | 0.93 | 1159 (1050, 1444) | 0.22 | 5.8 (4.5, 7.7) | 0.26 | 13.0 (8.7, 19.7) | 0.27 |
Outside | 43 (44.8) | 6.6 (5.8, 7.3) | | 0.07 (0.03, 0.12) | | 0.41 (0.28, 0.53) | | 1160 (1005, 1427) | | 5.8 (4.4, 7.7) | | 15.0 (10.5, 25.3) | |
Home located near |
Main road6 |
5 m or less | 32 (32.3) | 6.7 (5.5, 8.2) | 0.22 | 0.07 (0.04, 0.13) | 0.66 | 0.44 (0.31, 0.60) | 0.73 | 1135 (982, 1416) | 0.60 | 6.0 (4.6, 8.1) | 0.46 | 17.1 (9.7, 27.1) | 0.10 |
¼ mile + | 67 (67.7) | 6.5 (5.7, 7.2) | | 0.09 (0.04, 0.14) | | 0.41 (0.28, 0.54) | | 1195 (1032, 1429) | | 5.7 (4.4, 7.7) | | 13.1 (9.0, 20.2) | |
Landfill6 |
No | 94 (94.9) | 6.6 (5.7, 7.3) | 0.57 | 0.09 (0.04, 0.13) | 0.88 | 0.41 (0.28, 0.54) | 0.01 | 1165 (1009, 1429) | 0.87 | 5.7 (4.4, 7.4) | 0.30 | 13.6 (9.3, 21.6) | 0.65 |
Yes | 5 (5.1) | 6.5 (6.5, 6.8) | | 0.06 (0.04, 0.11) | | 0.76 (0.39, 1.2) | | 1160 (1049, 1208) | | 7.9 (6.4, 8.9) | | 13.0 (12.5, 24.3) | |
Fuel station6 |
No | 89 (89.9) | 6.6 (5.7, 7.3) | 0.97 | 0.08 (0.04, 0.13) | 0.71 | 0.42 (0.27, 0.55) | 0.37 | 1171 (1038, 1429) | 0.88 | 5.8 (4.5, 7.7) | 0.70 | 14.7 (9.7, 24.1) | 0.12 |
Yes | 10 (10.1) | 6.7 (6.0, 7.3) | | 0.10 (0.02, 0.13) | | 0.39 (0.35, 0.45) | | 1055 (965, 1260) | | 4.7 (3.9, 6.6) | | 9.6 (8.6, 12.9) | |
Child’s School |
Attends school? |
No | 84 (84.0) | 6.6 (6.8, 7.3) | 0.64 | 0.08 (0.03, 0.12) | < 0.01 | 0.42 (0.29, 0.54) | 0.37 | 1206 (1030, 1437) | 0.93 | 5.8 (4.4, 7.3) | 0.15 | 13.6 (9.4, 23.8) | 0.94 |
Yes | 16 (16.0) | 6.4 (5.2, 6.9) | | 0.12 (0.08, 0.20) | | 0.40 (0.27, 0.51) | | 1100 (1018, 1232) | | 6.9 (4.9, 9.7) | | 12.9 (8.5, 24.6) | |
Cigarettes |
Smoke in home6 |
No | 83 (83.8) | 6.6 (5.8, 7.4) | 0.25 | 0.08 (0.03, 0.13) | 0.31 | 0.41 (0.28, 0.58) | 0.77 | 1195 (1009, 1429) | 0.69 | 5.8 (4.4, 7.7) | 0.34 | 13.5 (9.4, 20.9) | 0.94 |
Yes | 16 (16.2) | 6.3 (5.3, 6.8) | | 0.08 (0.06, 0.13) | | 0.43 (0.30, 0.50) | | 1124 (1041, 1429) | | 6.0 (4.9, 7.7) | | 15.0 (8.3, 26.6) | |
More than 96% of children lived in houses with roofs made of iron sheets; 97% of households used charcoal as the main cooking fuel; and none of the 100 children ate canned food, making assessment of the contribution of these potential pollution sources to blood metal levels impossible. Similarly, no family members worked in a printing press, metal fabrication facility, quarry/mine, or at a battery or electronics recycling facility. One parent worked at a fuel station, one at a paint factory, and one at a metal smelting facility.
Discussion
We report a high prevalence of elevated blood levels of lead, manganese, antimony, cobalt, cadmium, and copper in a community-based sample of otherwise healthy Ugandan children living in the Katanga urban settlement of Kampala. Of concern, nearly all children had elevated blood lead and one in three had elevated blood manganese. Each of these metals has established neurobehavioral consequences, necessitating further study of the sources of exposure.
Strengths of the study include the large number of heavy metals assessed, the sampling methods used, and the environmental information collected. Although the sample is representative of the children living in Katanga, the homogeneity of the sample with regard both to common environmental exposures and the universally high values of certain metals made identifying environmental determinants difficult. Although nearly all children had elevated blood lead, none of the predictors on our questionnaire was associated with higher blood lead levels. Lead was phased out of gasoline in Uganda in 2005 [
10]. One survey of blood lead among Ugandan children living near the Kiteezi landfill 12 km north of Kampala also reported a high prevalence of elevated blood lead, with 20% of children, compared to 10% in our study, having a value above 10 μg/dL [
10]. Proximity to the landfill was the strongest predictor of blood lead. Water from community and home taps has been associated with high blood lead in Uganda [
11], but in our study, blood lead was not significantly different among children whose drinking water came from community taps, home taps, or well/spring. Although there are currently no legally-binding controls on lead paint in Uganda [
12], having painted walls, doors, or windows was also not associated with blood lead levels.
In addition to small numbers in some categories, the observed lack of difference among these categories may have stemmed from the fact that lead was universally high. Blood lead greater than 2 μg/dL has been associated with deficits in IQ, cognitive delay, and behavior [
13], underscoring the necessity of identifying the sources of exposure in this population. A 2008 study in Uganda reported that water from Lake Victoria, the source of much of the tap water in Kampala, was well above the WHO threshold for lead in drinking water [
11]. Soil and plants were also high in lead in the same study, with higher lead content found among plants and soil adjacent to main roads as compared to 2 km away from main roads. We did not find a difference in lead among children living adjacent to a main road versus one-quarter mile away from the main road in our study. Follow-up work that directly tests the lead content of water, air, and soil near the Katanga settlement is needed to identify the sources of exposure.
More than one third of children had high blood manganese. Manganese is an essential trace mineral, but excess manganese in children has been associated with deficits in IQ, increased risk of ADHD, and poorer verbal function [
14]. Drinking water is a commonly identified source of toxic levels of manganese [
15], but water source was not significantly associated with manganese concentrations in our study. Instead, living in a home with cement floors or cement walls was significantly associated with higher blood manganese as compared to walls or floors made of mud or clay. Manganese is a common ingredient in cement [
16], and manganese has been reported to be in cement dust from factories in the United States [
17], but cement as a cause of high blood manganese is not commonly reported in the literature. Further testing of the manganese content of the cement used in the Katanga settlement is needed.
All children had an elevated antimony level, although no environmental exposure measured predicted higher antimony levels. Antimony alloys are commonly used in batteries, sheet metal, solder, bearings, castings, ammunition, and pewter [
18]. When any of these materials are smelted or recycled, antimony is released into the air. Some of this antimony then returns to the soil with rainfall, highlighting a potential source of contamination. Car exhaust is another commonly reported source of antimony exposure [
19].
The five children who lived near a garbage landfill had significantly higher blood cobalt than children not living near the landfill, and the 16 children who attended school had significantly higher blood cadmium than children who did not attend school. The significance in these metal levels despite the small number children with these exposures underscores the importance of these sites, i.e., the landfill and school, as potential sources of significant heavy metal exposure. Cobalt is found in magnets, grinding and cutting tools, paints, batteries, and cobalt-coated metal from electroplating of electronics [
20]. Chronic cobalt exposure can cause asthma, decreased pulmonary function, and lung fibrosis [
20]. Cadmium is released into the environment from a wide range of sources, including tobacco smoking, waste incineration, and fossil fuel combustion [
21]. Thus, the possible sources of exposure to each of these metals are plentiful in Katanga and in the broader Kampala area.
Like manganese, small amounts of copper are required by the body, but excess exposure can cause nose and throat irritation, vomiting and diarrhea, and liver and kidney damage [
22]. Potential sources of copper are plentiful and include exposure to waste water, water from copper pipes, cooking with copper cookware, or copper-containing fungicides. It is critical to note that children in Katanga have high exposure to waste water, as there is an open drainage sewer channel in the settlement. Twelve percent of the children in our sample had a high blood copper concentration, but blood copper concentration did not differ by levels of any of the environmental factors assessed. The effects of prolonged copper exposure on child neurobehavioral development are unclear.
Our pilot study was limited by small sample size and by a sample that had similar exposures which potentially limit the applicability to other populations. Certain trends, such as increased blood lead in children who drank well/spring water and higher antimony in children whose houses had windows may have reached statistical significance with a larger sample size. Also, because some exposures, such as roofs made of iron sheets and household cooking done with charcoal, were common to nearly all children, the association of these key heavy metal sources to blood levels of metals could not be assessed in this initial study.
Nevertheless, our community-based study revealed that otherwise healthy Ugandan children have significantly high blood levels of six heavy metals. Excess exposure to each of these metals is associated with harmful health consequences, and high levels of some, notably lead and manganese, are established contributors to impaired neurobehavioral development. These neurodevelopmental deficits are increased synergistically with exposure to mixtures of multiple heavy metals. Thus, testing of the soil, air, cement, water, and paint in the Katanga settlement for metal content is needed, as are nutritional studies. Approximately 40% of Ugandan children of this age are iron-deficient [
23]. A high prevalence of iron deficiency among children in the current study may up-regulate absorption of many of the divalent metals, including lead and manganese, as these metals would all be absorbed through the intestinal transport protein divalent metal transporter-1, which preferentially absorbs iron, but also transports other divalent metals [
24]. Follow-up longitudinal studies with nutritional and neurobehavioral assessment before and after remediation of the identified sources of metals would demonstrate the burden of metals exposure to the health, nutritional status, and long-term development of children living in Katanga.