Lead poisoning remains a major public health problem, particularly for young children in developing countries. Several epidemiological investigations have shown a high prevalence of elevated blood lead concentrations among Bangladeshi children living in large urban and industrial centers [
This research was conducted in Sirajdikhan, located in the Munshiganj District in Bangladesh, a primarily rural and agricultural area largely devoted to the cultivation of rice and potatoes. Sirajdikhan is located 29.3 km southwest of Dhaka city, with a total population of 1.4 million and population density of 1,439 persons per km [
Participants were children who are currently enrolled in an ongoing population-based prospective birth cohort study designed to investigate the association between prenatal arsenic exposure and neurodevelopment in children. Participants were recruited from areas in which Dhaka Community Hospital (DCH) operates rural health clinics. After high blood lead concentrations were identified during an ongoing study, we selected 30 participants for environmental exposure assessment. Participants were grouped into quartiles according to blood lead concentrations, and 15 subjects from both the highest and lowest quartiles were selected for participation through random digit assignment. Twenty-eight families agreed to participate in exposure assessment activities.
Details of the recruitment strategy, eligibility criteria, and sample collection from pregnant women and newborns have been published previously [
Lead concentrations were measured in blood samples collected from fingerstick blood collected at 20 to 40 months of age, with a subset confirmed with samples obtained via venipuncture. Fingerstick blood lead concentrations were measured using portable LeadCare II instruments (Magellan Diagnostics, Billerica, MA, USA) that have a reportable range of 3.3–65
Blood samples obtained via venipuncture were collected in trace metal-free tubes for measurement of arsenic, manganese, and lead concentrations, the aims of the primary cohort study. Samples were analyzed for lead at the trace metals laboratory at HSPH in Boston, Massachusetts. Blood samples were first weighed (~1 g) and digested for 24 hours in 2 mL of concentrated nitric acid. These samples were then treated with 1 mL of 30% hydrogen peroxide per 1 g of blood and left overnight. Samples were subsequently diluted to 10 mL with deionized water. Lead concentrations were measured using a dynamic reaction cell-inductively coupled plasma mass spectrometer (DRC-ICP-MS, DRC II, Perkin Elmer). Analyses were performed using an external calibration method using lutetium as an internal standard to account for instrument drift. Continual calibration standards and the use of a standard reference solution (NIST 1643e: Metals in Water) were used to monitor precision and accuracy of the analysis. The final sample concentration used in the results was an average of 5 replicate measurements for each individual samples.
Quality control included analysis, procedural blanks, duplicate samples, spiked samples, and analysis of a certified reference material (NIST 955b: bovine blood for lead) to monitor for contamination, accuracy, and recovery rates. Recovery rates for lead in quality control and spiked samples were 81%–108%, and precision was measured as % relative standard deviation (SD), with a result of less than 3% for lead. The average limit of detection (LOD) was 0.2
Trained study staff administered questionnaires that collected medical histories and demographic information. This questionnaire included detailed questions about potential sources of lead exposure including paint, distance to roads and shops, battery recycling, parents’ occupations, and children’s hand-to-mouth activity.
Site visits were conducted at each participant’s home, and trained staff collected soil, drinking water, turmeric, and rice samples in trace metal-free 50 mL plastic centrifuge tubes. Three battery recycling/recharging plants were visited, and soil samples were collected from a region of soil closest in proximity to the shop. Soil samples were extracted from an approximately
Soil samples were dried and filtered through a 200
Lead bioaccessibility in turmeric samples was estimated using the simple bioavailability extraction test (SBET), an in vitro gastric fluid extraction that simulates metal dissolution in the stomach. The SBET was performed following procedures previously established by the US Environmental Protection Agency (EPA) [
We calculated descriptive statistics for both blood lead measures and selected subject characteristics. We dichotomized blood lead concentrations at different reference levels corresponding to current US and World Health Organization guidelines (i.e., ≥5
In this analysis, we used data from 309 children who provided fingerstick blood samples as of May 1, 2013. Venous blood samples were available for 176 (57%) of these children at the same visit. This sample represents 74% of the planned enrollment from this site for the birth cohort study. Table
Selected characteristics of study population.
Sirajdikhan, Munshiganj District, |
|
---|---|
Mean (SD) | |
|
|
Age (years) | 2.4 (0.2) |
BMI (kg/m2) | 16.7 (2.4) |
Hematocrit |
35.6 (3.9) |
|
|
Number (%) | |
|
|
Male sex | 158 (51.1) |
Place of birth | |
Home | 115 (37.2) |
Clinic or hospital | 194 (62.8) |
Mother’s education | |
No primary education | 31 (10.0) |
Primary education | 119 (38.5) |
Secondary education | 149 (48.2) |
Any higher secondary | 10 (3.2) |
Father’s education | |
No primary education | 57 (18.4) |
Primary education | 134 (43.4) |
Secondary education | 102 (33.0) |
Any higher secondary | 16 (5.2) |
Eating nonfood items (pica) | |
Yes | 146 (47.5) |
No | 162 (52.4) |
The overall median fingerstick blood lead concentration at approximately the age of 2.5 years was 8.1
Percentage of children at or above selected blood lead concentrations.
Blood lead concentration ( |
Fingerstick ( |
Venous ( |
---|---|---|
≥5 | 78.3 | 84.1 |
≥10 | 26.5 | 25.6 |
≥20 | 1.6 | 1.0 |
Lead concentrations varied among our turmeric samples, with a mean of ~80
Lead concentration in turmeric samples shown with total lead in turmeric and the fraction that is bioaccessible in simulated gastric fluid. Samples with lead concentrations below the limit of detection were excluded from bioaccessibility analyses.
Almost 80% of children aged 20 to 40 months tested in Munshiganj had blood lead concentrations at or above 5
In our study, turmeric samples had high concentrations of lead, with high levels of bioaccessibility, suggesting that contamination of food spices may be an important source of lead exposure in this setting. The elevated levels of lead in these turmeric samples (mean, ~80
Previous reports of elevated blood lead concentrations in Bangladeshi children have focused on large cities and industrial areas [
Previous research has identified lead-contaminated foodstuffs originating from Bangladesh and India, further supporting our turmeric findings. A recent study by Saha and Zaman [
While this research identifies the presence of lead-contaminated turmeric, the method of contamination is unknown. Uptake of lead from soil into the turmeric is a possible, but unlikely, source of contamination, as previous studies estimate the maximum uptake of lead into the root of the plant to be approximately 10% [
We were not able to assess the amount of turmeric ingested by children, and this is a further limitation of this study. We did not find a strong correlation between turmeric lead concentrations and blood lead concentrations, though we note that the turmeric samples were obtained months after the blood concentrations were measured. We therefore report that the high levels of lead in turmeric are a potential risk to the community.
There is an urgent need to identify the source of lead exposure in rural communities in Bangladesh. Turmeric, a spice commonly used in Bangladeshi cooking, is a potential source of lead exposure. Further testing is needed to confirm these sources. Once identified, preventative measures can be taken.
This study was approved by the Human Research Committees at the Harvard School of Public Health (HSPH) and Dhaka Community Hospital (DCH).
The authors report no conflict of interests.
The authors would like to thank study staff at Dhaka Community Hospital for performing all field work, including subject recruitment, specimen collection, and quality control. The study staff at Dhaka Community Hospital includes Md Shariful Islam Shakil, Ronjit Halder, Azizul Islam Razu, Sakila Afroz, Imam Hossain, Salim Mia, Monowar Hossain, Abul Kalam, Afroza Khatun, Abdul Kader, Nazmul Huda, Nur Islam Tareq, Amirul Islam Tutul, Israt Jahan, Aminul Islam, and Mostofa Kamal. Ema Rodrigues supervised laboratory quality control in Boston. Fareesa Islam and Nancy Diao provided assistance with data management. Shaye Moore and Elizabeth Jarvis provided editorial assistance. This study was funded by the Department of Environmental Health, Harvard School of Public Health. Dr. Mazumdar was supported by K23 ES017437 from the National Institutes of Environmental Health (NIEHS), National Institutes of Health (NIH). Additional support was provided by NIEHS Grants P42 ES16454 and P30 ES000002.