Measurement Uncertainty of Ethanol Concentration in Venous Whole Blood Determined By a HS-GC-MS Method

Since the alcohol intake strongly affects the driving capability and the Blood Alcohol Concentration (BAC) is related to the car crash risk, many countries define threshold values of BAC for drivers. In this work the measurement uncertainty of the BAC was estimated in view of the use of the BAC value in the legal procedures. Moreover, the in-house validation of the HS-GC-MS analytical method, used for the determination of the BAC, was carried out to ensure a suitable level of quality of the measurement. The method employed uses atrixmatched references for both calibration and quality control procedures. The parameters examined during the validation procedure were: sensitivity, linearity and uncertainty of the calibration, limit parameters, carry over, precision and trueness. Both intra-assay repeatability and intermediate precision were evaluated at 0.5 and 0.8 g L-1 of ethanol. In the second case, the variability of three factors, i.e. operator, liquid dispenser and time, was considered. The limits of detection and quantitation are 5.8 ∙ 10-4 and 1.8 ∙ 10-3 g L-1, respectively. The values of intra-assay precision at 0.5 and 0.8 g L-1 are 1.7% and 2.2% g L-1, respectively. Those of intermediate precision are 6.7% and 5.6%. The method provides unbiased results. Various contributions were taken into account to assess the uncertainty budget by the bottom up approach. Relative combined standard uncertainty, for the two concentration levels (in bracket), are: ucr (0.5)=3.4% and ucr (0.8) = 3.1%, using intra-assay repeatability, and ucr (0.5) = 7.3% and ucr (0.8) = 6.0% with intermediate precision.