Evaluation of measurement uncertainty of dissolution tests by the ISO-GUM approach and Monte-Carlo simulation

Abstract

Measurement uncertainty, no doubt, is an important element in examining and interpreting the results obtained in chemical analyses, particularly in the pharmaceutical field. Obviously, this is required by several guides and standards; however, few documents are interested in estimating the measurement uncertainty associated with dissolution test results. In this paper, the Guide to the expression of Uncertainty in Measurement (GUM) and its Supplement 1 "Monte-Carlo Simulation" (MCS) were used, side by side, to estimate the measurement uncertainty of two dissolution test methods, each using a different instrumental technique, namely UV-Visible Spectrophotometry and High Performance Liquid Chromatography (HPLC). A comparative study of both approaches was successfully conducted then with the aim of not only examining the compatibility between the results of the two methods, but also the presentation and discussion of the advantages and limitations of the applicability of each approach in drug analyses. The results obtained with the ISO-GUM and the Monte-Carlo simulation, for the two cases of analysis, revealed comparability between the estimates of uncertainty. Indeed, the standard uncertainties obtained by the two approaches are very close (uGUM Repaglinide (RG) = 1.1885% & uMCM Repaglinide (RG) = 1.1854% and uGUM Irbesartan (IB) = 1.4028 % & uMCM Irbesartan (IB) = 1.3071%). On the other hand, it has been found that the ISO-GUM approach slightly overestimates the expanded uncertainty because of the used value k = 2 of the coverage factor. Moreover, we have found that the conditions of applicability of the analytical approach and its numerical complement are not always obvious, in particular, when the model of calculation of the measurand is complicated and the instruments of measurements used are complex.