The effects of chronic exposure to a food contaminant depends mainly on the amount consumed (the dose response effect). Thus, food consumption data and the degree of contamination of the food by the contaminant in question are required for a reasonable estimation of chronic exposure. This study aims to fill this gap in information in this important area of public health.
The authors of this study adopted the approach of a “total diet study” method which is the recommended basis of exposure estimation to the contaminants in question. Further, it was carried out in a CKDu hotspot. A dietary survey coupled with food analysis was carried out. Blood lead was studied in the same exposed population to further confirm the finding of exposure estimation, as it was the heavy metal of highest concern in the study.
Diet survey
It was found that rice, vegetables and inland fish comprise the major dietary components in the study population. The average consumption of rice as estimated is higher than the published district and national per capita average consumption levels as well as the WHO G14 cluster of global dietary consumption group for cereals. Rice was also found to comprise the largest component by weight in the diet in the study cohort. It should be noted that the district and national averages are derived from studies which include the entire population, which include children and the elderly, who consume smaller quantities, while the subjects in this study were mostly CKD patients who were also over 40 years of age. Further, the subjects admitted to consuming less food since they fell ill (their sickness tended to depress appetite) and admitted to consuming greater quantities when they were well. The authors believe that the uniquely high consumption levels of rice in the study population offer a credible basis to embark on further related studies aimed at deriving potential causation of CKDu from chronic exposure to heavy metals. Vegetable consumption levels appear to be marginally lower than both the district and national averages.
Commonly consumed vegetables were fruit vegetables, pulses, roots/tubers, starchy vegetables, green leafy vegetables and, Leguminous pods.
The survey revealed very low consumption levels (60 g) of inland fish per week in this population. Allinson G et.al and Premarathna HL et.al, also have previously described that freshwater fish consumption is not a crucial dietary component in the study population [
34‐
36].
Males were found to consume significantly higher amounts of inland fish than their female counterparts (p = 0.017). There was no significant statistical difference between consumption of rice and vegetables between males and females.
The main source of income in the study cohort is agriculture; and the resultant low income, poor infrastructure facilities and rural living conditions may serve to explain the amounts and limited range of food types by this population.
The study population was found to consume food mostly from their own farms (65%) or food purchased from nearby village markets which sell produce from nearby farms, and which may in effect contain similar levels of toxic heavy metals. This finding may indirectly support the observed geographical localization of the Chronic Kidney Disease (CKDu hot spots). Further it can be argued, if the source of the contamination of the food is not aggressively controlled, there is danger that the adverse health effects of those contaminants may spread from its boundaries and in due course cause similar adverse effects on the wider population of Sri Lanka, including the urban regions.
Determination of model diet and exposure assessment
For the exposure assessment, methodology used by Joint WHO/FAO committee on food additives (JEFCA) and European Food Safety Authority (EFSA) panel on Contaminants in the Food Chain (CONTAM) was adopted.
For the purpose of calculation of chronic total exposure, our diet survey and national data available from other sources were taken in to consideration and a model diet for 60 kg man was presumed: A diet of 3 kg of rice with 1 kg of vegetables per week was assumed to be consumed by a healthy man weighing 60 kg living in the study area as the model for exposure calculation. Mean lead, cadmium and arsenic levels in the food commodities consumed by the subjects are used for the calculation of exposure of the people living in the area.
If the rice and vegetable lead values were taken in to consideration a 60 kg man is exposed to 236.3 μg × 3 kg + 215.2 μg x1kg = 924.1 μg of lead per week. Contribution from other food varieties, tobacco smoking, water, air, dust and soil might conceivably add a few additional micrograms of lead to this value.
Effect of above exposure was further established by the blood lead study of exposed subjects; Forty two percent (42%) of study population had detectable levels of lead in their blood (LOQ 3.3 μg/dL). The value ranged from 1.7 (extrapolated) to 8.8 μg/dL with a mean of 3.0 μg/dL
+ SD1.7. Similar blood lead levels ranging from 1.03 to 9.09 μg/dL (minimum and maximum respectively) with a mean of 3.6 μg/dL were found in a comparative study population in Medirigiriya in Sri Lanka [
6].
Until 2011, JECFA considered the Provisional Tolerable Weekly Intake (PTWI) for lead as 25 μg/kg body weight (bw), which is equivalent to 1500 μg for a 60 kg person. However, in 2011, JECFA reported that a PTWI of 25 μg/kg bw was associated with a decrease of at least 3 IQ points in children and an increase in systolic blood pressure of approximately 3 mmHg in adults; and concluded that the PTWI of 25 μg/kg bw can no longer be considered as being appropriate to protect health. This resulted in the previously recommended PTWI [
11] being withdrawn.
The mean chronic exposure to lead in the study populations is estimated as 62% of the pre-2011 JECFA PTWI, pointing to insufficient protection from lead. As this is a mean value it can be further concluded that in a certain segment of the study population, chronic exposure to lead is a matter for greater concern considering the recent scientific conclusions of health effects of such exposures. The detected lead levels are known to cause decreased IQ levels in children as well as increasing systolic blood pressure and reduction of GFR (Glomerular Filtration rate) in adults, which increases the public health significance of the finding. If the JECFA limit of lead exposure is applied to the study population, it can be assumed that something close to 3 IQ points decrease in children and close to 3 mmHg systolic blood pressure increase can be predicted in the study population.
The detection of blood lead in the subjects consuming contaminated food types strongly suggest the presence of bioavailable lead in food. Lead is a known heavy metal toxin that crosses the placenta and blood-brain barrier, depositing in foetal tissues. The evidence on neurological consequences of prenatal exposure to lead appears to reflect changes in cognitive impairment. Primary prevention of lead exposure across the entire human life span is being recommended in current scientific literature [
16].
The phenomenon of fluoride being known to increase the absorption of lead may be playing a synergistic role in this situation as it had been reported that fluoride concentration in water in the CKDu affected areas is higher [
37].
The European Food safety Agency (EFSA) Panel on Contaminants in the Food Chain (CONTAM) considers developmental neurotoxicity, systolic blood pressure increases and chronic kidney disease as the most potent adverse effects of lead [
14]. Dose response modelling for these adverse effects was used to determine the 95th percentile lower confidence limit of the benchmark dose (BMD) of 1% extra risk. The respective BMDLs derived from blood lead levels and corresponding dietary intake values for developmental neurotoxicity, BMDL
01, are 1.2 μg/dL and 3.50 μg/kg body weight per week respectively. If these values are applied to the population of our study, they would be predicted to result in a lowering of 2.5 and 4.3% of full-scale Intelligence Quotient (IQ) when the average blood lead levels and corresponding dietary intake is considered respectively. Here it is assumed that exposure levels of children are similar to adults, when in fact exposure of children may actually be higher due to their lower body weight, thereby placing them at a higher risk than estimated.
The CONTAM panel has determined the 95th percentile lower confidence limit of the benchmark dose (BMD) of 1% mean annual increase of population based systolic blood pressure BMDL01, as a blood lead level of 3.6 μg/dL, which corresponds to dietary intake values in μg/kg body weight per week of 10.50 μg. If these values are applied to the population of this study, levels of 1.2 and 1.5% annual increase of population based systolic blood pressure can be derived. Generally, such an increase in populations is associated with a significant risk of stroke and other cardiovascular events.
The CONTAM panel has determined the 95th percentile lower confidence limit of the benchmark dose (BMD) of 10% change in the prevalence of CKD defined by GFR below 60 mL/ 1.73m2 body surface per minute, BMDL10, as 1.5 μg/dL for blood lead and a corresponding dietary intake of lead equal to 4.41 μg/kg body weight per week. If these values are applied to the population of our study, they are at risk of 20% and 36% reduction in GFR respectively when the average blood lead levels and dietary intake of lead exposure is considered.
Retarded neurodevelopment in children of Sri Lanka as a consequence of exposure to lead emerges as a serious concern as the average Intelligence Quotient (IQ) of the Sri Lankan population is 79, which positions it below 131 countries according to the IQ Research [
38].
In 2016 low birth weight (LBW) was recorded at 16%, underweight among children less than 5 years old at 20.5%, wasting at 15.1% and stunting at 17.3% in Sri Lanka [
32]. Ischemic heart disease and neoplasms are among the 10 leading causes of hospitalization and death [
39]. Thirty percent of the 35 to 64-year-old Sri Lankans are hypertensive [
40]. All these indicators point to significant health concerns with consequential societal, economic and other adverse effects. In addition to the above health effects, a number of studies have shown that lead and cadmium exposure is associated with low birth weight [
17].
The mean total cadmium exposure levels determined in the model diet is 22.0 μg × 3 kg of rice + 17.7 μg × 1 kg of vegetables = 83.7 μg per week. Other food, water and other sources may contribute an additional exposure to cadmium amounting to a few micrograms. The corresponding tolerable weekly intake (PTWI) is 152 μg per week for a 60 kg man [
15]. This indicates that the study population is within the accepted tolerable limits when mean exposure from food only considered.
However, previous studies have shown higher concentrations of cadmium: [
1‐
8]. A number of possible explanations can be given for this observed discrepancy, namely: the samples have been collected at different periods of time and from different sources, different analytical methods may have been used, and there may be seasonal variation in contaminants found in the commodities. In our study food samples were collected from affected households where the sampled food was retained for their own consumption. In contrast, some of above studies used samples sourced from the market.
The main health effects of chronic exposure to cadmium is nephrotoxicity, malignancy and association with low birth weight. Cadmium also appears to be associated with overall cancer mortality in men and women [
10,
18,
19]. Multivariate models have established relationships between blood cadmium tertials and foetal growth parameters; namely birth weight, low birth weight, birth weight percentile by gestational age, small for gestational age, pre-term birth, length, and head circumference [
20‐
22].
Chronic total exposure of arsenic appears to be of 45.5 μg X 3 + 43.8 μg × 1 = 180.3 μg per week in a 60 kg man. Provisional tolerable weekly intake (PTWI) recommended for arsenic by JECFA until 2011 was 15 μg per kg body weight which is equal to 900 μg for a 60 kg man. The study population is exposed to 20% of the tolerable intake showing arsenic exposure in the study population seems not to pose a significant health risk. However, in 2011 JECFA has withdrawn the PTWI of 15 μg per kg body weight stating that this guideline is no longer appropriate, as such exposure falls within the Benchmark dose (BMDL) of a 0.5% increase in incidence of lung cancer [
41]. However, as the estimated exposure of arsenic is five times less in the study population, it can be assumed safely that the increase in incidence of lung cancer may be very much lower than 0.5% from arsenic exposure through food in our study population.