Validation of Analytical Methods Based on Chromatographic Techniques: An Overview

Abstract

The purpose of any analytical method is to provide consistent, reliable, and accu-rate data. For this reason, the performances and the limitations of the method, as well as the external influences which may modify these features, must be deter-mined prior to its use. Validation plays a major role in achieving this goal [1,2]. The word " validation " comes from the Latin term validus, meaning worth/strong, thus suggesting that something is true, useful, and reliable [3]. The most accurate definition of validation is that provided by ISO 9000:2000 as the confirmation, by means of a thorough examination and obtaining realistic and unequivocal evidences, that the procedure is effectively applicable for its indented purpose [4]. Validation is the act of proving that any approach, strategy, experimental pro-cedure, process, laboratory staff, instrumentation, reagents, and room conditions selected for the method will function in a proper way under a fixed set of condi-tions. Besides, it can be used to individually evaluate the appropriateness of these factors [5]. The validation evaluates the range and conditions of applicability, and checks if every future measurement in routine analysis will provide a con-centration of the analyte close enough to the true value [6]. In addition, it can also quantify the degree of coincidence of a measured concentration and the true value, by the calculation of the bias and the uncertainty associated with the result [7]. Therefore, the validation verifies if the method is suitable to be used as a quality control tool and for research support [8]. It is an essential step in method development, which must be implemented by laboratories to prove they can produce analytical data with high reliability [9]. The validation consists in the determination of well-defined quality parame-ters: statistical (selectivity, specificity, calibration curve, linearity, calibration range, accuracy, precision, recovery, uncertainty, limit of detection, limit of quantification (LOQ), decision limit, detection capability, robustness, stability, system suitability, and comparison with other methods) [1,2,10] and operating/ economical (cross contamination, simplicity, analysis time, price per analysis, safety for laboratory staff, and environmental impact) [11,12]. The results from