Low-level chronic mercury exposure in children and adolescents: meta-analysis.

Background: Mercury is a well-known neurotoxin. There are three kinds of mercury exposure: elemental mercury poisoning, inorganic mercury poisoning and organomercury poisoning. Organomercury is the most toxic. Twenty-four hour urine for mercury and blood mercury are the gold standards for diagnosis of mercury poisoning, including low-level chronic mercury exposure. Other tests for mercury level are discussed. The purpose of the present paper was to review recent data on the nature, pathophysiology, pharmacokinetics, diagnostic methods, treatment and the linkage to neurodevelopmental disabilities of mercury exposure in children.

Methods: A literature search was undertaken of MEDLINE (1980-2003), and American Academy of Pediatrics, American Medical Association, American Dental Association, World Health Organization and Center for Disease Control websites. The search string 'mercury' was used in MEDLINE and articles were selected as appropriate by two independent reviewers. All relevant information was reviewed and data were extracted by two independent reviewers.

Results: Based on the meta-analysis of the accuracy of hair mercury, hair mercury levels correlated with mercury level in blood (sample size weighted correlation coefficient, r w = 0.61), with 24 h urine ( r w = 0.46) and with cord blood ( r w = 0.64). However, the correlation for hair mercury level with 24 h urine level and blood level was not high enough to replace them in clinical decision-making of individual patient. Epidemiological evidence has shown that low-level mercury poisoning is not a cause of autism (relative risk = 0.49, 95%CI = 0.36-0.66). The risk of neurodevelopmental disabilities from low-level exposure to methylmercury from the regular consumption of fish is still controversial even after combining results from different epidemiological studies worldwide. There is a lack of data in the literature about the effect of chelation therapy in children with neurodevelopmental disabilities.

Conclusion: Mercury poisoning should be diagnosed only with validated methods. There is no evidence to support the association between mercury poisoning and autism.