As the world of medical technology progresses, so does the landscape of regulatory expectations. Gone are the days when unknown chemicals would be allowed or accepted in preclinical testing. For today’s regulatory bodies, ambiguity in chemical characterization is no longer tolerated. Put simply, unknowns are unacceptable. This approach requires a risk-based approach, demanding a holistic view of all chemical constituents or, in some cases, a detailed focus on a singular compound of concern.
Whether expected or unspecified, all compounds from extractables and leachables (E&L) testing must be analytically screened. ISO 10993-18:2020 explains that analytical screening aims to “discover, identify and semi-quantitatively estimate the concentration of all relevant analytes in a test sample above an established reporting threshold.” Analytical testing labs must first adequately discover or detect the extracted compounds to begin the identification or semi-quantitation processes.
Common compound identification methods
Liquid chromatography-mass spectrometry (LC-MS) is commonly used in E&L testing to identify and semi-quantitate semi-/non-volatile compounds. However, there are several detection techniques for these types of compounds. One approach is to utilize an ultraviolet (UV) detection system, which is limited in detecting compounds without a light-absorbing moiety.
Alternatively, mass spectrometry (MS) detection has emerged as the leading option for detecting semi-/non-volatile compounds in E&L testing. In addition to not relying on a compound with light-absorbing functionality amid other advantages, detection by MS offers more thorough results and complete toxicological risk assessments.
LC-MS vs. UV detection
There are numerous advantages to detecting organic compounds by mass spectrometry during chemical characterization of medical device extracts. Generally, more chemical compounds may be observed by MS detection due to a broader range of chemical functionality that is amenable to ionization in the mass spectrometer (e.g., various carbonyl-containing structures, amines, polyethers, and siloxanes).
In addition, typical LC-MS experiments are performed by analyzing both positive and negative ion modes to increase the range of compounds observed by exploiting different means of ionization. This can further be extended to use both electrospray ionization (ESI) as well as atmospheric-pressure chemical ionization (APCI), yielding up to four unique datasets by which to detect and identify non-volatile and semi-volatile compounds (e.g., ESI+, ESI-, APCI+, APCI-).
The detection limit for most chemical compounds is also lower in MS analysis versus UV (Figure 1). This can allow the analyst to identify more compounds of potential toxicological relevance to help ensure the most robust risk assessment of the medical device. Furthermore, in the event of LC-UV-MS analysis, the MS data is being collected regardless. So, excluding compounds in the characterization report based on UV signal alone will arbitrarily remove compounds from identification and the toxicological risk assessment.
The primary limitations to UV-based detection of semi-volatiles and non-volatiles of extractables and leachables of medical devices are coverage of chemical functional groups and detection limits. Many compounds extracted from medical devices are UV-silent or provide weak signals, such as aliphatic compounds containing esters, amines, ethers, and siloxanes.
In general, much higher concentrations are required to detect samples by UV versus signal from a mass spectrometer. To achieve these higher concentrations, additional sample concentration is often necessary to arrive at the study’s target analytical evaluation threshold (AET), which can exacerbate issues involving compound degradation, volatility, and sample matrix effects.
Detection by UV can also fail to account for significant portions of the non-volatile residue (NVR) mass performed for exhaustive extractions. Similarly, potential compounds of toxicological concern may be present in a relevant amount but below the limit of detection by UV.
Figure 1: UV and MS detection of Irganox 1098 with concentrations ranging from 0.05 μg/mL to 10 μg/mL. The lowest concentration with a signal-to-noise ratio greater than 3 by UV was 0.5 μg/mL.
A case study of a silicone-containing nitinol stent
After performing NVR analysis, it was determined that there was approximately 350,000 μg of non-volatile extractables from an isopropanol extraction per device. Examining the LC-UV-MS data by UV-detection showed no discernable peaks at 200 nm wavelength, and expanding this to detect in a range from 190-400 nm also revealed no peaks.
Conversely, detection by ESI-MS in both positive and negative ion modes revealed numerous peaks (Figure 2) that were further identified as various cyclic siloxanes, fluorinated siloxanes, amines, polyethers and plasticizers. During toxicological risk assessment, these fluorinated siloxanes were determined to have margin of safety (MOS) values significantly below 1.
Figure 2: Overlay of exhaustive extractables of a silicone-containing nitinol stent. Top: MS detection in positive-ion mode, Middle: MS detection in negative-ion mode, Bottom: UV detection at 200 nm wavelength
A final word
As medical device testing undergoes rapid evolution, the quest for thorough chemical characterization becomes paramount. While historically significant, UV detection faces undeniable limitations, especially concerning functional group coverage and sensitivity.
The prowess of LC-MS is undeniable. It casts a broader net, capturing a wide range of compounds at lower concentrations and providing intricate layers of analysis through its diverse ionization modes. The case of the silicone-containing nitinol stent illuminates the sheer contrast between UV and LC-MS detections, emphasizing the risks of potential undetected hazards. However, harnessing the full potential of LC-MS requires expertise and nuanced understanding. Medical device manufacturers who might not possess this specialized know-how in-house stand at a pivotal crossroads. Here, the role of an experienced lab testing partner becomes invaluable. Teaming up with a seasoned laboratory ensures that the complex methodologies of LC-MS are navigated with precision while also assuring comprehensive chemical characterization and thorough toxicological risk assessments.
