a systematic review and meta-analysis of case-control and cohort studies
The association between childhood leukemia and extremely low frequency magnetic fields (ELF-MF) generated by power lines and various electric appliances has been studied extensively during the past 40 years. However, the conditions under which ELF-MF represent a risk factor for leukemia are still unclear. Therefore, we have performed a systematic review and meta-analysis to clarify the relation between ELF-MF from several sources and childhood leukemia. We have systematically searched Medline, Scopus, Cochrane Database of Systematic Review and DARE to identify each article that has examined the relationship between ELF-MF and childhood leukemia. We have performed a global meta-analysis that takes into account the different measures used to assess magnetic field exposure: magnetic flux density measurements (<0.2 µT vs. >0.2 µT), distances between the child’s home and power lines (>200 m vs. <200 m) and wire codings (low current configuration vs. high current configuration). Moreover, meta-analyses either based on magnetic flux densities, on proximity to power lines or on wire codings have been performed. The association between electric appliances and childhood leukemia has also been examined. Of the 863 references identified, 38 studies have been included in our systematic review. Our global meta-analysis indicated an association between childhood leukemia and ELF-MF (21 studies, pooled OR=1.26; 95% CI 1.06–1.49), an association mainly explained by the studies conducted before 2000 (earlier studies: pooled OR=1.51; 95% CI 1.26–1.80 vs. later studies: pooled OR=1.04; 95% CI 0.84–1.29). Our meta-analyses based only on magnetic field measurements indicated that the magnetic flux density threshold associated with childhood leukemia is higher than 0.4 µT (12 studies, >0.4 µT: pooled OR=1.37; 95% CI 1.05–1.80; acute lymphoblastic leukemia alone: seven studies, >0.4 µT: pooled OR=1.88; 95% CI 1.31–2.70). Lower magnetic fields were not associated with leukemia (12 studies, 0.1–0.2 µT: pooled OR=1.04; 95% CI 0.88–1.24; 0.2–0.4 µT: pooled OR=1.07; 95% CI 0.87–1.30). Our meta-analyses based only on distances (five studies) showed that the pooled ORs for living within 50 m and 200 m of power lines were 1.11 (95% CI 0.81–1.52) and 0.98 (95% CI 0.85–1.12), respectively. The pooled OR for living within 50 m of power lines and acute lymphoblastic leukemia analyzed separately was 1.44 (95% CI 0.72–2.88). Our meta-analyses based only on wire codings (five studies) indicated that the pooled OR for the very high current configuration (VHCC) was 1.23 (95% CI 0.72–2.10). Finally, the risk of childhood leukemia was increased after exposure to electric blankets (four studies, pooled OR=2.75; 95% CI 1.71–4.42) and, to a lesser extent, electric clocks (four studies, pooled OR=1.27; 95% CI 1.01–1.60). Our results suggest that ELF-MF higher than 0.4 µT can increase the risk of developing leukemia in children, probably acute lymphoblastic leukemia. Prolonged exposure to electric appliances that generate magnetic fields higher than 0.4 µT like electric blankets is associated with a greater risk of childhood leukemia.
In summary, our study suggests that exposure to ELF-MF higher than 0.4 µT increases the risk of developing leukemia in children. Acute lymphoblastic leukemia is probably the subtype of leukemia associated with ELF-MF. Prolonged exposure to electric appliances that generate magnetic fields higher than 0.4 µT like electric blankets is associated with a more elevated risk of childhood leukemia. The distance from power lines linked to leukemia is difficult to determine but living more than 200 m away from power lines is likely a safe distance for children not associated with a higher leukemia risk.
This work was supported by the research grant from Electric Power Research Institute. SETIL was financially supported by research grants received from the Italian Association on Research on Cancer (AIRC), the Ministry for Instruction, University, and Research, the Ministry of Health, the Ministry of Labour, and Piedmont Region.
Zhang Y, Lai J, Ruan G, Chen C, Wang DW. Meta-analysis of extremely low frequency electromagnetic fields and cancer risk: a pooled analysis of epidemiologic studies. Environ Int. 2015 Dec 15;88:36-43. doi: 10.1016/j.envint.2015.12.012.
• A significant association between ELF-EMF exposure and cancer risk was identified.
• Subgroup analysis revealed increased risk only in North America, especially in United States.
• However, the data from individual European country was contradicted with each other.
• Increased risk was only observed in residential exposure or interview-based surveys.
• Device measured studies obtained no significant association in overall effects.
Human-made electromagnetic fields (EMFs), such as nonionizing radiation, are classified into three categories: extremely low frequency fields (< 300 Hz) ....
Studies have suggested that extremely low frequency electromagnetic fields (ELF-EMF) may affect physiological functions in animal models. However, epidemiologic studies investigating the association of ELF-EMF with the susceptibility to cancer yield contradictory results.
In this comprehensive analysis, we conducted a search for case-control surveys regarding the associations of ELF-EMF and cancer susceptibility in electronic databases. A total of 42 studies involving 13,259 cases and 100,882 controls were retrieved.
All studies were case–control studies, including 23 breast cancer ..., 1 testicular cancer ..., 1 acoustic neuroma ..., 1 endometrial cancer ...., and 3 other cancer cases.
Overall, increased susceptibility to cancer was identified in the ELF-EMF exposed population (OR=1.08, 95% CI: 1.01, 1.15, P=0.02). In the stratified analyses, increased risk was found in North America (OR=1.10; 95% CI: 1.02, 1.20, P=0.02), especially the United States (OR=1.10; 95% CI: 1.01, 1.20, P=0.03). However, studies from Europe contradict these results. Moreover, a higher risk was found to be statistically significantly associated with the residential exposed population (OR=1.18; 95% CI: 1.02, 1.37, P=0.03). Furthermore, an increased cancer risk was found in interview-based surveys (OR=1.16; 95% CI: 1.00, 1.35, P=0.04). In device measurement-based studies, a slight increased risk was found only in premenopausal breast cancer (OR=1.23; 95% CI: 1.01, 1.49, P=0.04).
Our meta-analysis suggests that ELF-EMFs are associated with cancer risk, mainly in the United States and in residential exposed populations. Methodological challenges might explain the differences among studies.
The overall evaluation conducted by the World Health Organization (WHO) indicated that extremely low-frequency magnetic fields (ELF-EMFs) are possibly carcinogenic to humans (Group 2B), while static electric and magnetic fields and extremely low-frequency electric fields are not classifiable as to their carcinogenicity to humans (Group 3) (Humans, 2002).
ELF-EMFs are mainly generated by power transmission lines, power equipment or appliances (Chen et al., 2013). Because of the rapid development of industry and society, humans are surrounded by various electric devices, and exposure to ELF-EMFs is increasing. Currently, the biological effects induced by ELF-EMFs on human health have become a cause for concern (Grellier et al., 2014 and Zhang et al., 2015).
In general, animal experiments have produced positive results for all known human carcinogens, for which adequate testing has been performed (Humans, 2002). However, it is notable that childhood leukemia is the only cancer outcome for which this association has been consistently found using epidemiological methods (Grellier et al., 2014). It has been hypothesized that experiments designed following the classical two-step initiator–promoter concept of carcinogenesis may not be appropriate for understanding the biological effects of ELF-EMFs, as disease progression may result from complex interactions of genotoxic and non-genotoxic carcinogens (Juutilainen, 2008).
The results in Fig. 2 show that weak association between EMF-ELF exposure and susceptibility to cancer was identified when all the eligible studies were pooled (OR = 1.08, 95% CI: 1.01, 1.15, P = 0.02) regardless of the exposure models or cancer types (Fig. 2).
In the country subgroup analysis, a statistically significant increase in risk was found in North America (15 breast cancer, 3 brain cancer, 1 leukemia and 1 other cancer; OR = 1.10, 95% CI: 1.02, 1.20, P = 0.02), mainly in the United States (14 breast cancer, 3 brain cancer and 1 other cancer; OR = 1.10, 95% CI: 1.01, 1.20, P = 0.03). On the contrary, no statistically significant association between EMF-ELFs and cancer risk was found in a global analysis of European studies (7 breast cancer, 7 brain cancer, 8 leukemia and 5 other cancers) .... An increased risk of cancer was found in Norway (3 breast cancer, 1 brain cancer, 1 leukemia and 1 other cancer; OR = 1.11, 95% CI: 1.02, 1.21, P = 0.02) and France (1 brain cancer and 1 leukemia; OR = 1.38, 95% CI: 1.03, 1.84, P = 0.03), while a decreased risk was found in Sweden (4 breast cancer, 1 brain cancer, 1 leukemia and 1 other cancer; OR = 0.90, 95% CI: 0.84, 0.96, P = 0.001) and England (2 brain cancer, 2 leukemia and 2 other cancers; OR = 0.82, 95% CI: 0.69, 0.96, P = 0.02). In addition, a study from New Zealand also showed an increased cancer risk (1 leukemia; OR = 1.97, 95% CI: 1.08, 3.59, P = 0.03) .... Further subgroup analyses based on cancer type did not reveal any statistically significant associations in all of the analyzed types. When compared by exposure methods, an increased risk was only observed in residential exposure populations (OR = 1.18; 95% CI: 1.02, 1.37, P = 0.03).
[Note that the 23 studies that examined breast cancer risk yielded a marginally significant association with ELF-EMF exposure (OR = 1.07, p = .06)].
In conclusion, in our study, relevant literature selected from broad databases with stringent standards revealed an increased risk of cancer upon ELF-EMF exposure. However, more quantitative studies will contribute to more comprehensive results in the future.