Comparison of hair, nails and urine for biological monitoring of low level inorganic mercury exposure in dental workers
Creatinine-corrected urine mercury measurements in spot urine samples are routinely used in monitoring workers exposed to inorganic mercury. However, mercury measurement in other non-invasive biological material has been used in some epidemiological studies. Dentists and dental nurses remain a group of workers with potential exposure to inorganic mercury through their handling of mercury-containing amalgam, although changes in work practices have reduced the current, likely exposure to mercury. Therefore, dental workers remain an occupational cohort in whom the value of using different biological media to identify exposure to low level inorganic mercury can be investigated. Samples of head hair, pubic hair, fingernails, toenails and urine were analysed for mercury content from a cohort of UK dentists (n=167) and a socioeconomically similar reference population (n=68) in whom any mercury exposure was primarily through diet. The mercury content in all biological material was significantly higher in the dental workers than in the control population (p<0.0001). The geometric mean and 90th percentile mercury concentrations in the urine samples from dentists were 1.7 and 7.3 μmol mol-1 creatinine, respectively, with only one sample having a value at around the UK's Health and Safety Executive biological monitoring health guidance level of 20 μmol mol-1 creatinine. Receiver operator characteristic analyses suggested that the ability of the biological material to discriminate between dentists and referents were fingernails>urine≈toenails>pubic hair≈head hair. Further investigation is warranted as to why fingernails appear to be such a good discriminator, possibly reflecting some contribution of direct finger contact with amalgam or contaminated surfaces rather than systemic incorporation of mercury into growing nails. Good correlation between head hair and pubic hair mercury levels in all subjects was obtained (r=0.832), which was significantly improved when hair samples weighing <10 mg were excluded (r=0.868). Therefore, under these exposure conditions and using the described pre-analytical washing steps, there is little influence from atmospheric contamination on the level of mercury content of head hair. The choice of non-invasive biological materials for mercury analysis depends on a number of considerations. These include the toxicokinetics of urinary mercury excretion, the growth rates of hair and nail, the nature and time-frame of exposure, and the fact that urine mercury may not reflect the body burden level from dietary methyl mercury. However, the data from this study suggests that urine mercury remains the most practical and sensitive means of monitoring low level occupational exposure to inorganic mercury.