Optimization of ICP-MS internal standardization for 26 elements by factorial design experiment.

INTRODUCTION Internal standardization is a common tool used in inductively coupled plasma - mass spectrometry (ICP-MS) analysis in order to reduce matrix effects, and thus improve the reliability and the robustness of results. However, its efficacy relies on the choice of a proper internal standard (IS) which, ideally, undergoes the same signal variations as the analyte. Thus far, IS selection has mainly relied on the proximity of atomic mass between the analyte and the internal standard. However, while it may be a satisfactory rule of thumb, more recent works suggest that this criterium might not be suitable in several conditions, among which the presence of high amounts of carbon atoms in the sample. This may thus be of particular interest in the case of trace elements determination in biological samples. MATERIALS AND METHODS In this study, we propose an empirical and global approach to IS selection in ICP-MS through the use of a factorial design of experiments (DoE), with a focus on biological matrices of interest in clinical analysis: human blood and urine. The suitability of 13 potential IS was evaluated for 26 clinically-relevant analytes, including a polyatomic ion obtained through reaction with oxygen, across 324 experimental conditions. RESULTS AND DISCUSSION The results underline several exceptions to the rule of IS selection based on mass proximity, notably when considering heavy or polyatomic analytes. As a consequence, measurements of said analytes in several extreme experimental conditions using IS selected by mass proximity could yield vastly erroneous results (up to 30 times the theoretical concentrations). By contrast, the use of empirically selected IS yielded much more acceptable results.