Analysis of microsamples of biological fluids via inductively coupled plasma-tandem mass spectrometry using a micro-flow injection system coupled to a high-efficiency sample introduction setup

Abstract

In many fields, the sample volume available is the limiting factor for ultra-trace elemental analysis, and thus, the development of novel methods that can extract more information from lower sample amounts is a topic that deserves attention. In this work, the performance of a micro-flow injection system coupled to either (1) a traditional or (2) a high-efficiency sample introduction system has been evaluated for the analysis of liquid microsamples using inductively coupled plasma-mass spectrometry (ICP-MS). The figures of merit with each sample introduction system have been studied in a wide range of sample volumes and sample uptake rates. The use of a high-efficiency low-volume sample introduction system provided a > 4-fold enhanced sensitivity as compared to the traditional one, while the former setup was demonstrated being capable of accurately delivering volumes as low as 5 or 10 μL to the ICP. Even for such low sample amounts, good linearity (>0.9998) and repeatability (RSD down to 0.2%) were experimentally demonstrated, while matrix effects were lower with the high-efficiency sample introduction system. The use of an internal standard adequately corrected for changes in sample introduction conditions, allowing quantitative results to be obtained by using a single set of calibration data obtained at fixed operating conditions (sample volume and sample uptake rate), with a bias <15% (typical acceptance criterion for clinical QC analysis). The set-up characterized was successfully applied to the analysis of microsamples of biological fluids (down to 0.4 μL of sample volume). Tandem ICP-MS (ICP-MS/MS) was used to overcome spectral overlap (chemical resolution with a mixture of NH3/He) affecting the determination of six clinically relevant metals (Co, Cr, Mn, Ni, Ti and V) at ultra-trace concentration levels. No significant differences were found between the results obtained in this work and the reference values, even when using different sample introduction conditions for the measurement of calibration standards and samples.