Comparison of approaches for quantifying extractables in a variety of polymeric medical device materials with LC-MS/MS

Quantification by LC-MS using electrospray ionization (ESI) is challenging due to various factors that influence signal intensity. The gold standard involves the use of isotopically labelled standards spiked into the sample. However, for the analysis of extractable compounds from polymeric materials in medical devices, such standards are often unavailable. Nevertheless, these extractables must be monitored and quantified in accordance with regulatory requirements. In this study, we compared different LC-MS-based approaches for quantifying extractables from polymers commonly used in medical devices. The evaluation focused on 25 analytes, partially of high regulatory relevance under the European Medical Device Regulation. Two primary strategies were investigated: quantification using internal isotope-labelled surrogates, and the use of relative response factors (RRFs) from an experimentally established database. Using the closest eluting internal surrogate resulted in a mean quantification error of over threefold, which improved to 2.7-fold when surrogates were selected based on structural similarity for ESI+ mode while no significant improvement was observed for ESI- mode. In contrast, employing the RRF database yielded a lower bias, with a mean error of 1.6-fold (ESI+) and 2.3-fold (ESI-) using structurally similar compounds and as low as 1.2-fold when the analyte's specific RRF was available. These approaches were applied to 22 polymer materials spanning 8 polymer types. Each was extracted using three different solvents and the impact of matrix effects on quantification accuracy was assessed. Although matrix effects such as ion suppression were observed, their influence was minor compared to the differences among the quantification approaches.