Trace Metal Determinations in Biological Samples

As analysts, we often forget and/or fail to communicate to our customers that the numbers which are actually generated from an analysis are used to predict the concentration(s) of analyte(s) which is (are) present in the bulk material. As such they are estimates only and are subject to uncertainties. These uncertainties can be appreciable. For a trace element analysis to be of any value whatever, two conditions must be rigidly met. The analytical sample, which is actually presented to the instrument (typically less than 1 mL) must be (i) homogeneous and (ii) a miniature replica of the bulk material which has been sampled. 

From a review of the repeatability of a broad range of analyses performed by the U.S. Food and Drug Administration (FDA) Horrwitz observed a correlation between the CV (coefficient of variation for replicate determinations) and the analyte concentration which was independent of (i) the analyte identity, (ii) the analytical method and (iii) the sample type. Roughly, a CV of 10% was to be anticipated if the analyte was present at the low mg/kg level, but 30-60% if the analyte was present at low pg/kg concentrations (Horwitz, W., and Albert, R., Anal. Proc. (London), 24, 49, 

The analysis of foods, and biological materials in general, for trace elements poses special analytical problems which are not encountered with other sample types. Certain elements of interest are typically present at levels ranging from low to sub pg/kg at one extreme whereas other analyte elements are present at more than 100 mg/kg. Since an analyte trace element can be 

A further major source of error is contamination trace elements are ubiquitous and unless special precautions are taken, the sample will inevitably be contaminated. It would seem that an attention to detail and efforts to avoid contamination which boarder on the fanatical is the only way to minimise contamination. As a rule of thumb, the total analytical reagent blank for a given trace element should be reproducible and should result in values which are at least a factor of ten lower than the lowest level being measured in a sample. 

---

The advent of inexpensive computing power has changed dramatically the way voltammetric determinations are performed. In particular, automated multielement determination schemes and high speed flow systems are now available for convenient and effective trace metal determination in biological samples. 

---