Cleaning verification is a critical process for patient safety in pharmaceutical manufacturing in order to keep cross-contamination below acceptable limits. A common cleaning verification method is the total organic carbon (TOC) swab. Others have studied the variances of different factors on the TOC swab in order to establish the best swab method. This paper attempts to quantify the sampling error of the TOC swab in the actual sampling situation using simulation. The study investigates the variability on the drug product recovery due to different analysts, concentration, steel finish and position, as well as the estimation of the given swab area. The results demonstrate that the sampling error leads to a large variation in TOC results. For areas estimated in the laboratory, it leads to an increase in limit of detection, LOD, with 60%, while for areas estimated in a tank, the LOD cannot be determined due to the large heteroscedasticity. Thus, this paper is also to be considered as an invitation to discuss and further investigate the TOC sampling error in the pharmaceutical industry.
Keywords: Monte Carlo simulation; Pharmaceutical cleaning verification; Total organic carbon swab.
Copyright © 2022 The Authors. Published by Elsevier B.V. All rights reserved.