Click on the links below to access all the ArticlePlus for this article.
Please note that ArticlePlus files may launch a viewer application outside of your web browser.
In pooled analyses of epidemiologic studies, Ahlbom et al1 found that exposure to time-weighted average power-frequency magnetic fields greater than 4 mG is associated with a 2-fold increase in the risk of childhood leukemia. For fields greater than 3 mG, Greenland et al2 found a 1.7-fold increase in risk. These findings are unlikely to be explained by chance, and confounding appears not to account for the whole of the association. However, selection bias has been proposed to be at least partially responsible for the association in case–control studies.3 This relation is difficult to interpret because of uncertain mechanism of action of power-frequency magnetic fields (the aspect of the fields that is biologically relevant) and the uncertain time period that would be etiologically relevant.
Magnetic fields of extremely low frequency do not appear to initiate cancer. Researchers have hypothesized that they may act as cancer promoters or progressors,4,5 possibly enhancing genetic change caused by known genotoxic agents.6 Therefore, for exposure to magnetic fields to play a role in the development of acute leukemia in children, it is possible either that the children must be genetically predisposed to cancer or that they be previously sensitized by exposure to cancer initiators. Children with Down syndrome may represent such a susceptible group with their 20-fold greater risk for developing acute leukemia.7 The aim of this study was to assess whether, among children with Down syndrome, an association exists between residential exposure to magnetic fields and the risk of developing acute leukemia.
From 1995 to 2003, a case–control study was performed with 42 cases of children with both Down syndrome and acute leukemia and 124 controls consisting of healthy children with Down syndrome. Children had to reside in Mexico City and be under 16 years of age.
In Mexico City, there are both public and private institutions that treat children with acute leukemia. The private institutions care for fewer than 5% of children with cancer.8 Of the 9 public institutions that treat children with cancer and that were invited to participate in this study, only 4 had cases from Mexico City. All cases with acute leukemia and Down syndrome during the study period were included.
In Mexico City, children with Down syndrome may be found in only 2 types of institutions. The 76 Centers of Multiple Attention (Centros de Atención Múltiple) provide special education to children with various disabilities; during the study period, 1409 children with Down syndrome were registered at these institutions. None of these Centers was included in the study because they did not include a karyotype diagnosis for Down syndrome. Five specialized centers required karyotyping as a requirement for admission; this study included 2: Instituto John Langdon Down and Centro de Atención Integral del Niño con Síndrome de Down. These centers accept children from any part of Mexico City. If a child at one of these 2 educational institutions were to have developed acute leukemia, one of the hospitals serving as sources of the cases would have taken care of the child. Of the 200 children at the Instituto John Langdon Down during 2002–2003, 103 were under 16 years of age; of 28 children in Centro de Atención Integral del Niño con Síndrome de Down, 21 were under 16 years of age. The 3 specialized centers not included had 166 children registered with 98 under 16 years of age. Thus, of the total number of children under 16 years of age in the specialized centers, 56% were included in this study.
The protocol of the study was approved by the Ethics and Investigation Committee of Instituto Mexicano del Seguro Social (No. 97-718-0006/98-718-0033). The parents of each child signed an informed consent form.
For both cases and controls, trained nurses conducted an individual, in-person interview of each parent of the child. Information included birth weight of the child (>2500 g and ≤2500 g); child's sex; mother's age when pregnant (>35 years and ≤35 years); family history of cancer; socioeconomic status (not crowded, ≤1.5; crowded, >1.5 persons per room); and place of residence in Mexico City (south of the city, which is an agricultural zone, vs the center–north of the city, where there is a high concentration of factories). Exposure at home to herbicides, fertilizers, or insecticides was classified as “yes” or “no.” Density of road traffic (≤2556 vehicles, >2556 vehicles per day, in the street with heavy traffic not more than 500 m from each residence) was determined. Details of the measurement of the variables are provided in the supplemental information available with the online version of this article.
A gaussometer (dosimeter; EMDEX II; Electric Field Measurements, West Stockbridge, MA) was used to take spot measurements of the magnetic fields. Measurements were taken for at least 5 minutes at the front door of the residence with the average value being reported.
On a standardized diagram (supplementary online Fig. 1), trained personnel (who had not been informed whether a residence belonged to a case or control) recorded the type of power lines and their distance to the residence. Three categories of exposure, as classified by Kaune and Savitz9 from the code of power lines, were analyzed (Supplementary Table 1). The code of power lines was validated for the houses of Mexico City (Supplementary information, validation).
To evaluate possible selection bias in the controls, hospital controls were also included in the study. Although children with acute leukemia are attended only in tertiary care hospitals, a child must be referred from a secondary care hospital. Hospital controls were selected from the secondary care hospitals that served as sources of the cases. Children from secondary care were preferred over those from primary care, because primary care patients come from a very restricted zone, thus increasing the possibility of overmatching of the environmental exposure to magnetic fields. Of 473 children, we included as controls only those (126 [27%]) who were admitted for surgery, who were without cancer or apparent malformations, and who lived with both biologic parents. We compared hospital controls with Down syndrome controls to determine 1) if their demographic characteristics were similar and 2) if any differences in the demographic characteristics were differentially related to exposure to magnetic fields. The variables tested were type of residence (owned, rented, other), the level of crowding, type of housing (house or apartment), number of rooms (1, 2, or more), and the type of floor (sealed or unsealed).
Statistical analysis was performed by calculating odds ratios (ORs) and 95% confidence intervals (CIs). Unconditional logistic regression analysis was performed to control confounding variables.
Cases were similar to controls for most characteristics with the exception of socioeconomic status, birth weight, and family history of cancer (Table 1). The median age for the cases was 90.5 months and for the controls, 84 months. There were 34 cases of acute lymphoblastic leukemia and 8 cases of acute myelogenous leukemia.
Direct Measurement of Magnetic Fields
Using logistic regression analysis, the risk of acute leukemia elevated with exposure to ≥6 mG magnetic fields (OR = 3.7; 95% CI = 1.05–13.06) (Table 2).
Configuration of Overhead Cable Near a Residence
Using logistic regression analysis, there was increased risk of acute leukemia with both medium exposure (OR = 5.81) and high exposure (OR = 4.05) (Table 2).
Evaluation of Possible Selection Bias
Sociodemographic characteristics were similar between the Down syndrome controls and the hospital controls. The only difference was in the type of flooring; Down syndrome controls had a higher frequency of sealed flooring than did hospital controls (63% vs 45%). However, the geometric mean of magnetic fields was always higher for the Down syndrome controls than for the hospital controls (1.49 mG and 0.95 mG, respectively) (Supplemental Table 2). With respect to the wiring configuration, no differences were found between Down syndrome controls and hospital controls (Supplemental Table 3).
To our knowledge, this is the first study to assess the effect of exposure to magnetic fields in relation to acute leukemia in children with Down syndrome. Controls had a more crowded households suggesting higher socioeconomic status for controls. However, compared with the hospital controls, Down syndrome controls did not show differences with respect to the crowdedness index or to housing conditions (except for flooring type). Additionally, Down syndrome controls all had higher exposure levels to magnetic fields than hospital controls, which could suggest an underestimation of risk.
In this study, only spot measurements of magnetic fields were collected; we did not have the opportunity to compare spot measurement with 24-hour measurements in the child's bedroom. However, in studies in which this comparison has been performed, the spot measurement has been shown to underestimate values compared with time-weighted average.11
We do not know if the spot measurement of magnetic fields at the homes of children correlated with measurements from the years before the interview or, for the cases, from the years before the disease. However, in another study,12 spot measurements performed on the same house 6 or 7 years apart had a correlation of 0.7.12
Our results support the possibility that exposure to magnetic fields ≥6 mG may be associated with the development of acute leukemia in children with Down syndrome, suggesting that genetic susceptibility to leukemia may modify an effect of magnetic fields.5,13,14
The authors thank Sylvia García of Fundación John Langdon Down; Susana Ramírez Robles, María de los Ángeles Rojas Ramírez, and Pedro González Vivanco of Fundación CTDUCA; María del Carmen Mejía of the Instituto CEDAC; A. Ishikawa of the Hospital de Petróleos Mexicanos; and O. Del Ángel-Guevara and F.J. Miganjos-Huesca of the Hospital Militar for their expertise. The authors thank Ana María Olivera of the Comunidad Down A.C. for supplying valuable information; and Guadalupe Vargas, Juan Carlos López, José E García, and Vicente Guadarrama of the Dirección de Educación Especial of the Seretaría de Educación Pública for information on the Centros de Atención Múltiple. The authors also thank their collaborators, Roberto Bernáldez-Ríos, Roberto Paredes-Aguilera, Antonio Ortiz-Fernádez, María Carmen Rodríguez-Zepeda, and Patricia Pérez-Vera for their participation in the capture of information on patients and comments on the manuscript; and María Carmen Martínez-García for supervising this investigation. The authors thank Veronica Yakoleff and Yolanda Castelazo for editing the manuscript.
1. Albohm A, Day N, Feychting M, et al. A pooled analysis of magnetic fields and childhood leukemia. Br J Cancer
2. Greenland S, Sheppard AR, Kaune WT, et al. A pooled analysis of magnetic fields, wire codes, and childhood leukemia. Childhood Leukemia-EMF Study Group. Epidemiology
3. Mezei G, Kheifets L. Selection bias and its implications for case–control studies: a case study of magnetic field exposure and childhood leukaemia. Int J Epidemiol
4. Crumpton MJ, Collins AR. Are environmental electromagnetic fields genotoxic? DNA Repair (Amst)
5. Feychting M, Ahlbom A, Kheifets L. EMF and health. Annu Rev Public Health
6. Cho YH, Chung HW. The effect of extremely low frequency electromagnetic fields (ELF-EMF) on the frequency of micronuclei and sister chromatid exchange in human lymphocytes induced by benzo(a
)pyrene. Toxicol Lett
7. Robison LL. Down syndrome and leukemia. Leukemia
. 1992;6(suppl 1):5–7.
8. Mejía-Aranguré JM, Fajardo-Gutiérrez A, Bernáldez-Ríos R, et al. Incidence trends of acute leukemia among the children of Mexico City: 1982–1991. Arch Med Res
9. Savitz DA, Kaune WT. Childhood cancer in relation to a modified residential wire code. Environ Health Perspect
10. UK Childhood Cancer Study Investigators.Exposure to power-frequency magnetic fields and the risk of childhood cancer. Lancet
11. Neutra RR, DelPizzo V, Lee GM. Evaluation of the possible risks from electric and magnetic fields (EMFs) from power lines, internal wiring, electrical occupations and appliances [California Public Utilities Commission (CPUC). California Electric and Magnetic Fields (EMF) Program web site]. June 2002. Available at: www.dhs.ca.gov/ehib/emf/RiskEvaluation/Chapter3.pdf
. Accessed April 20, 2006.
12. Savitz DA, Pearce N, Poole C. Update on methodological issues in the epidemiology of electromagnetic fields and cancer. Epidemiol Rev
13. Mejía-Aranguré JM, Fajardo-Gutiérrez A, Flores-Aguilar H, et al. Environmental factors contributing to the development of childhood leukemia in children with Down's syndrome. Leukemia
14. Röösli M, Künzli N. Magnetic field exposure and childhood leukaemia—moving the research agenda forward [Commentary]. Int J Epidemiol