Effects from environmental Mn exposures: a review of the evidence from non-occupational exposure studies.

OBJECTIVE: The risk posed to human health by environmental manganese exposure is unknown. Occupational-exposure outcomes may not extrapolate to environmental exposures due to the healthy worker effect and differences in dosage parameters which may affect the biological response. This paper attempts to combine the existing literature on non-occupational Mn exposures with results from our current study in SW Quebec on environmental Mn exposure (Mergler et al., this issue) within the framework of a biologically-based, dose-response (BBDR) model. BBDR MODEL: The basic BBDR model consists of seven stages relating exposure to health effects. The stages are: 1) sources, 2) applied dose, 3) absorbed dose, 4) target-site dose, 5) toxic event, 6) measurable change, and 7) health outcome.
RESULTS: Several air monitoring programs, such as the PTEAM study (Riverside, CA, 1990, mean PM10 Mn outdoor-airborne 24 h average = 0.045 microgram/m3), provided data relevant to the estimation of Mn applied dose, but did not include measures of body burden. Data from the SW Quebec study showed a mean total-particulate airborne Mn concentration of 0.022 microgram/m3 with a range of 0.009 to 0.035 microgram/m3 across four sampling sites, whereas the EPA reference concentration (RfC) is 0.05 microgram/m3. EPA has considered tap water levels to be safe below 200 micrograms/l Mn, and mean Mn tap-water (MnW) level in the participants' homes was 6.38 +/- 11.95 micrograms/l with a range from 0.1 to 158.9 micrograms/l Mn. A previous study of MnW exposure in Greece reported Mn levels in areas with low, medium and high MnW ranging from 4 to 2,300 micrograms/l and a significant association with Mn in hair but not Mn in blood (MnB). The mean absorbed dose of the SW Quebec study participants, as indicated by MnB, was 7.5 +/- 2.3 micrograms/l with a range of 2.5 to 15.9 micrograms/l. Our study and others on environmental Mn exposure did not provide an estimate of target-site dose. However, a significant correlation (r = 0.65) between MnB and signal intensity in T1-weighted MRI images has been reported in liver-disease patients with Parkinson-like signs who had MnB levels as low as 6.6 micrograms/l. Only animal and in vitro studies have provided evidence on the mechanisms of toxicity caused by Mn in the CNS. Several studies reported measurable changes in endpoints suggestive of a Parkinson-like syndrome in subjects with MnB levels ranging from 7.5 to 25.0 micrograms/l. Among other effects on neurobehavioral function observed in the current study was a significant relationship between MnB and the direction and speed of body-sway in men. The effects observed in these participants are sub-clinical and no health outcomes have been diagnosed. However, the Parkinson's disease incidence in the study area was previously reported to be 2-5 times higher than in the rest of Quebec, and several studies indicate that 25-35% of idiopathic Parkinson disease diagnoses are incorrect. Our study, the Greek study, and some clinical studies suggest that the risk of a Parkinson-like syndrome diagnosis may increase with continued Mn exposure and aging.
CONCLUSION: The limited data available for the BBDR model point to the need for evidence, particularly on relationships between Mn species, exposure route, MnB with chronic environmental exposure, ageing, and susceptibility factors, to improve human-health risk assessments for chronic, environmental Mn exposure.