Keeping our spirits up during a worldwide pandemic can be challenging. However, one bright spot on the horizon of our time is the increased invention and ingenuity that is coming into the medical device field to deal with the COVID-19 crisis. From 3D printed ventilators to new designs and materials for many mask types, there has been an explosion of new ideas and devices that are trying to make it to the market right now. However, with these new designs comes the challenge of how they will be assessed for safety by regulatory agencies. While many devices have received Emergency Use Authorization (EUA) for short-term use, supporting the long-term use of these devices becomes the subject in question. Therefore, along with new device ideas, new testing strategies need to be considered to correctly assess these new designs and ensure low patient risk.

For all medical devices, the endpoints to be assessed are stated in ISO 1099-1 and are also elucidated in the FDA guidance for Industry, Use of International Standard ISO 10993-1. However, gas pathway devices don't have specific guidance for endpoints to assess. While there is an ISO series (18526) that has testing strategies for gas pathway devices, it is challenging to utilize these testing guidelines with new devices and designs. When contemplating new testing strategies with gas pathway devices, there are really three areas that need to be considered when assessing the safety of these devices: Air volatile organic compounds (VOCs) and particulates, chemical characterization, and toxicological risk assessments.

Breathing Devices

Air VOCs and Particulates. For many breathing gas pathway devices, air VOCs and particulates can be difficult to assess. Many of the breathing gas pathway devices (ventilators, CPAPs, anesthesia machines, etc.) have different connections for the hoses, which many times the testing labs do not have adaptor for and, therefore, cannot direct the air to analyze to only the interior of the tubing. Assessing the whole device, interior and exterior, results in an increased level of air VOCs detected, as the outside surface of the device, which is not clinically relevant, is assessed in conjunction with the inside surface of the device. Being aware of this situation up front can lead to more appropriate testing. This issue can also be addressed by coordination with the testing laboratory and by providing appropriate connections so that only the interior surface will be tested. Another strategy is to use a surface area calculation with the assumption that the amount of air VOCs and particulates being shed can be scaled by the ratio of interior surface area to exterior surface area.

Chemical Characterization. Chemical characterization of breathing gas pathway devices has been a perplexing challenge. Questions arise regarding what to test, how to test it, and what data are needed to assess the device for safety in a way that makes scientific sense and ultimately assesses patient safety in the clinical setting. When conducting chemical characterization of breathing gas pathway devices or masks, the goal is to understand what compounds could be leaching from the device and reaching the patient. These leachables could contact the patient via condensate. Assessing this type of exposure is fraught with difficulties such as how to accurately extrapolate from a fully submerged device to what could contact the patient during clinical use.

For example, a mask that is connected to a ventilator will only have one side touching the patient's face. Because it is impractical to cap off the mask for extraction, the whole device must be submerged into the extraction solvent, thereby exposing materials that may not actually contact the patient to the solvent. Another difficulty is solvent selection for extraction. Would chemical characterization of the materials using strong, nonclinically relevant solvents correctly assess a patients’ potential risk, or would doing such work merely provide erroneous data that incites more chemistry work and would therefore prevent necessary devices from coming to market?

Toxicological Assessment. Solvent selection has been a hot button topic for all devices in extractables and leachables testing for quite some time, and breathing gas pathway devices have just furthered that discussion. With breathing gas pathway devices, water is typically used along with a secondary, semi, or nonpolar solvent. With water, the assessment is more straightforward as water is clinically relevant (as a simulant for humidified air). Other solvents prove to be difficult, as none of these devices are intended to be in contact with harsher semi and nonpolar solvents. As with the air VOCs and particulates discussion, extracting only the interior of the device in solvent is more problematic, since the ends of tubing or face masks cannot be blocked off to prevent solvent from escaping. This means that the whole device must be submerged in the extraction solvent, and an extrapolation from total surface area to interior surface area must be made to assess the clinically relevant concentration of the compounds that are detected.

More often than not, the adjusted data is not clinically relevant because it assumes that the total amount of solvent used for the extraction will be exposed to the patient. Often the volume of solvent used for extraction is in the 0.5–2.0-L range. But if clinical practices are considered, the amount of liquid that could potentially reach a patient is only the liquid condensation forming on the interior of the device tubing. A logical estimation for that volume of liquid is orders of magnitude below the volume of liquid used in the extraction testing. Understanding these challenges is key to properly testing and assessing these devices. Using this condensate example, there are ways to address this gap between how the device was tested and how the data can then be assessed to ensure patient safety. One way is to adjust the amount of extractable concentrations, which can be done in two ways — surface area or condensate volume — as discussed below.

The concentration of extractable compounds can be adjusted through consideration of the surface area extracted. The interior of the device is compared to the total surface area that was extracted to model the potential patient exposure concentrations against the total concentrations that were extracted. This method can often decrease the concentration of the extractable compounds by approximately 50 percent, if the exterior and interior are similar in surface area. However, this still is not an accurate approximation to the actual patient exposure.

A more accurate way to adjust the amount of extractable concentrations is to assess for a condensate volume. However, to do this, the amount of condensate needs to be determined. There are a couple of ways this can be accomplished, with the more common way being via gravimetric analysis. The device is run through a simulated test cycle with maximum humidification to determine the amount of water mass that can collect in the tubing. The device is weighed before and after the simulation, with the difference in weight being the amount of water condensate that has collected in the tubing during the simulation. A second option is to conduct a similar study but instead of a gravimetric weight determination, a drip chamber is used. The amount of water that collects in the drip chamber is assumed to be the maximum amount of condensate that could contact the patient. In using either measuring method, the concentration of extractables present is then decreased by the ratio of solvent volume in the extraction, to the total condensate volume that was determined.

Face Masks

Face masks are also an important aspect in the support of controlling the spread of COVID-19. While some testing was able to be addressed through justification for EUA approval, it is suggested that assessment for cytotoxicity, sensitization, and irritation be considered when assessing the safety of facial masks. There is a plethora of clinical data available to support that the masks, in general, are safe. When considering the biological approach to be taken for masks, it would be good to consider the draft guidance released from FDA in October 2020, “Certain Devices in Contact with Intact Skin.” This guidance, when fully approved, will allow for simpler justification of intact skin-contacting devices made from common materials with a history of safe use (i.e., polyester and cotton). One of the tests that can be problematic is cytotoxicity.

Many of the ear bands demonstrate a cytotoxic reaction due to the elastomeric content of the bands themselves. Options to address this outcome include to test the components of the mask individually or to perform a dilution series through a comparative study to show similar cytotoxicity to a face mask already approved and on the market.

Toxicological Assessment. The rationale for these approaches can be made in a toxicological risk assessment (TRA), which outlines how to assess the device in a way that accurately reflects the intended use and clinical application of the device. In the TRA, the aforementioned adjustments are used to understand what the clinically relevant exposure concentrations are and how to ultimately reduce any possible hazardous risks to the patients. These assessments must be completed using the best scientific thought, logical practices, and reasonings available to correctly ascertain any risk posed to the patient.

When presenting these arguments to any regulatory agency, it is imperative for both the composer of the risk assessment and the reviewer to use not only knowledge of the current standards, but also to understand the arguments and scientific reasoning used to assess the device and ensure patient safety. It is our duty, as engineers, scientists, and toxicologists to provide the best options available to patients and avoid getting caught up in a strict approach to testing in place of scientific reasonings.

This article was written by Christopher Pohl, Associate Toxicologist, Nelson Laboratories, LLC, Salt Lake City, UT. For more information, visit here .