PERFORMANCE ASSESSMENT IN FINGERPRINTING AND MULTI COMPONENT QUANTITATIVE NMR ANALYSES

The goal of this work was to set up a new quality control parameter suitable for performance assessment in multi component quantitative NMR analysis and NMR fingerprinting methods. In order to achieve the goal, an inter-laboratory comparison (ILC) was organized. 36 NMR data sets (corresponding to 1260 NMR spectra) were produced by 30 participants using 34 NMR spectrometers. A model mixture made up of five compounds [Aldicarb, Methamidophos, Oxadixyl, Pirimicarb and 3-(trimethylsilyl)-2,2,3,3-tetradeutero-propionic acid sodium salt (TSP)] dissolved in deuterated water was submitted to NMR analyses. The analytical target of the comparison was the quantification of analytes by the calibration line method. Such a method was chosen as it allows for identification of a theoretical line to be taken as reference in performance assessment. The NMR experiment considered for the interlaboratory comparison consisted of a single 90° excitation pulse preceded by a selective pre-saturation step. Nine NMR signals were selected for this study: three for Aldicarb, one for Methamidophos, two for Oxadixyl, two for Pirimicarb and the singlet of TSP which was taken as reference. Results show that quantitative NMR is a robust quantification tool. Performance assessment was carried out on single component quantification, by the popular and traditional z-score, and on multi-component analyses by means of a new performance index (named Qp-score) which is related to the difference between the experimental and the consensus values of the slope of the calibration lines. By an analogous reasoning followed for z-score, performance assessment by Qp-score is considered satisfactory when |Qp|<=2.0, questionable when 2.0<|Qp|<3.0 and unsatisfactory when |Qp|>=3.0. This study introduces a new quality control parameter, Qp-score, suitable for harmonization of fingerprinting protocols and quantitative multi component analysis. Such parameter, that was designed considering consolidated internationally agreed statistics, represents an unbiased evaluation tools for NMR method validations. Qpscore accounts for laboratory performance in terms of both instrumental adequacy and operator skill and enables laboratories to pooling of NMR data in suitable databanks. Moreover, Qp can be valuable for the development of multi-laboratory metabolomic platforms.