Systematic robustness testing of a liquid chromatographic method: A case study

Abstract

Robustness testing of a method plays a crucial role in establishing its reliability. It examines the potential sources of variability in one or more responses of the proposed method. In this study, the robustness testing of a method proposed for simultaneous determination of warfarin and its two process related impurities was evaluated by using two level, fractional factorial design. Factors that are sensitive to a variation during method transfer were selected as independent variables [aqueous content (range: 39-43%, v/v), concentration of acetic acid (range: 0.08-0.12%, v/v), flow rate (range: 0.93-1.33 mL/min), and wavelength (range: 218-222 nm)]. Variables that determine the quality of separation, viz., retention factor of the first peak, resolution between the critical peak pair, tailing factor of warfarin, and total analysis time were selected as responses. Robustness was assessed by graphical (half normal probability and Pareto plots) and statistical (analysis of variance) methods. It was found that, among the studied variables, aqueous content had a significant effect on capacity factor and analysis time. Furthermore, non-significant intervals for significant factors were established by contour profiling. This study demonstrated the significance of experimental design and other statistical tools in understanding the effects of investigating factors of the chromatographic system and in defining their limits.