Toenail clippings from females were examined over a period of approximately one year to investigate variability in temporal intra-individual toenail-Hg levels, and chronologically matched hair and toenail samples were compared to more accurately define the toxicokinetic variability of MeHg levels observed between the two compartments.
Hair-Hg levels from this population of females have previously been investigated with results showing no significant change in MeHg body burden levels across the study period suggesting that if MeHg population-based body burden levels were changing they were doing so slowly and across a time period larger than the one investigated . This study corroborates that conclusion and further, indicates that while there was temporal stability of the three toenail-Hg levels for the group as a whole, individuals had variable MeHg body-burden levels but the levels did not fluctuate by more than two-fold across the one-year period investigated. The fluctuation in individual body burden levels is likely due to changes in MeHg intake across sample intervals while intra-individual toxicokinetic variability may also be a contributing factor. The intra-individual change in Hg levels observed across the study period which did not exceed 1.5-fold for half of the participants and two-fold for nearly the entire study group can be considered normal behavioral variability that is non-directional fluctuation within longer term temporal stability. For individuals with stable body burden levels, a single data point value, such as toenail-Hg levels obtained at a given time, could be used to reflect long term MeHg body burden; with the individual’s mean value being within the interval defined by one-third to twice the obtained toenail-Hg level. It must be noted however, that for identifying body burden levels across a short time period, even within those having temporal stability, biomarker based measurements from a single time point used to estimate exposure can miss short term elevated MeHg exposure periods consisting of single exposures to those lasting weeks [36–38].
For toenails to be used as a biomarker for long term exposures such as years, there is a need to either verify temporal stability within the study population by comparing biomarkers reflecting body-burden levels against an empirically derived ratio as defined herein, or repeated biomarker determinations would be needed so as to understand long term body burden characteristics. If a population has stable MeHg body burden levels, or if long term averages can be identified, biomarkers such as toenail could be used to identify causal association even though the biomarker may be chronologically removed from disease manifestation. These determinants are relevant to the relationship between ongoing exposures and endpoints resulting from chronic exposure (i.e. cardiovascular disease) but would not be as applicable to circumstances associated with the population consisting of women of child bearing age. Long term variability will not be as relevant for gestational exposure as present day understanding suggests a finite window of susceptibility for women during pregnancy may exist that can result in deleterious effects; possibly even from a single exposure resulting in a large bolus dose at a critical time period during fetal development.
A limitation of this work is that the intra-individual temporal variability observed in toenail-Hg levels, even though only two fold, could be a factor leading to imprecision in exposure measurement. Estimates of imprecision for biomarker parameters such as hair and blood that can impact dose estimations in dose–response models have already been provided and the toenail compartment as a reservoir reflecting MeHg body burden may include similar imprecision . When investigating dose–response relationships, unbiased imprecision can lead to an underestimation of the relationship between dose and response. In the present work, where two toxicokinetic compartments are being compared, the imprecision will not lead to bias but will provide for a random error effect.
Chronologically matching hair with toenails provided for high correlation and yielded a slope of 2.79 based on a regression model (Table 1). Data that were obtained from samples collected at the same point in time (i.e. during a clinic visit) and thus not chronologically matched with respect to exposure period were also shown to correlate and were similar to results found in other studies comparing data from biomarkers collected at the same time but not representing the same period of exposure (Table 1). Nonetheless, the chronologically matched ratio should be used to estimate Hg levels in the hair compartment based on toenail levels. The chronologically matched ratio would allow for estimating the corresponding Hg levels in hair that have been widely used as a biomarker for investigating multiple toxicological effects, including cardiovascular outcomes and, would allow for comparison across studies as the hair-Hg data set within the literature is robust.
For any given population, if the hair- and toenail-Hg ratios from chronologically matched and unmatched samples are in reasonable agreement, we may conclude that their body burdens are relatively stable over time. From the existing population data sets (Table 1), this reasoning would suggest that the non-occupationally exposed Japanese women , one group of non-occupationally exposed Finns  and possibly the control group from within the dentists study  may have stable body burden levels. However, even if the body burden in a population is found to be, on average, stable over a given time period, it cannot necessarily be assumed that such stability can be extrapolated to much longer time periods. Thus, attempts at identifying stability in body burden within a study population should be made if biomarkers chronologically removed from disease manifestation are to be used to investigate a dose response relationship.