As anyone who has gone through the process can attest, ensuring medical device compatibility with the human body requires careful decision making across multiple, sometimes seemingly unrelated dimensions. Chicken-and-egg decisions on the testing sequence for endotoxins, sterility, biocompatibility, cleanliness, aging, packaging, and the production environment are common challenges.

To make matters more complicated, design changes to one factor may require retesting of not only that factor, but also retesting of different requirements entirely… or not!

This article will help to understand a number of common human compatibility factors and highlight some interdependencies between them in order to help determine how to approach initial testing sequences and change control. Each section will give a brief description of a specific factor related to human compatibility testing and any interdependencies of the previous sections.

To help those of us who like examples, I’ll use an imaginary device called the Gluco-Quick—which rapidly injects sugar into the bloodstream—to give specific examples. This is summarized at a high level in the chart below:

Bacterial Endotoxins

Bacterial endotoxins are toxic substances found in the cell walls of gram-negative bacteria. When these bacteria break down, endotoxins are released onto the medical device. If these are then carried into the human body—even in trace amounts—they can cause severe reactions such as fever, inflammation, or even septic shock. Thus, endotoxin testing is mandatory for devices that contact tissue or bodily fluids, such as surgical instruments, catheters, or implants, with the tightest limits being applied to cardiovascular and cerebrospinal fluid applications.

The standard method for endotoxin detection is the Limulus Amebocyte Lysate (LAL) assay, which helps ensure the levels of endotoxins are within safe limits. ANSI+AAMI ST72 is a widely accepted standard for endotoxin testing methods and limits, including by the U.S. Food and Drug Administration (FDA). It also has a nice bit of history and background that’s worth a read!

For our imaginary Gluco-Quick device, which has a syringe and needle built in, we would need to submit test samples to LAL testing regularly, both ahead of FDA clearance and periodically during production.

Sterility and Bioburden

Sterility is the process in which viable microbes are killed, often through exposure to steam, gases, or radiation. The first step is to assess the bioburden (that is, the number of viable microbes on an unsterilized device), then use that information to shape the sterilization process (e.g., the exposure time). Medical device sterility testing follows the ISO 11737 family of standards, which include validation methods and specific protocols based on device classification.

Though they’re related, sterility and bacterial endotoxins aren’t always 100% correlated: sterilization kills bacteria, but the bacteria don’t need to be alive for the endotoxins to be present on the device.

A device can be sterile after sterilization and still contain unacceptable endotoxin levels. Another factor is timing: devices that don’t support bacterial growth should have similar endotoxin levels whether or not they are sterilized a day or a year after assembly, but devices that support bacterial growth will see an increase in endotoxin levels until the device is sterilized.

Gluco-Quick is mostly plastic and metal and we wish to gamma sterilize it. First, we will measure the bioburden levels and then choose an appropriate level of radiation exposure, which we will then validate. If we can also prove the biotoxin levels are unaffected by sterilization timing, we can do our LAL tests any time before or after sterilization.

Biocompatibility

Biocompatibility is generally tested under ISO 10993-1, which evaluates how devices interact with biological systems. Simply put, it’s the expected interaction between the device materials and the human body, directly (i.e., through direct contact) or indirectly (i.e., through the transfer of chemicals into a fluid, which then contact the body). ISO 10993 has many elements and sub-elements and thus requires a number of tests to show compliance, unless existing proof can be found that exposed materials are known to be biocompatible.

There are also particular standards for biocompatibility of specific types of devices. For example, ISO 18562-1 covers breathing machine gases, particulates, and leachables.

Though not directly related, the materials that make up a medical device and the way they are processed can impact bacterial endotoxin levels and be affected by sterilization. Unless the materials in a device are known to be unaffected by sterilization, it’s a more conservative approach to test biocompatibility on sterilized devices for this reason.

As Gluco-Quick has a lot of plastic, we’ll be sure to test biocompatibility after sterilization. Bacterial endotoxins aren’t directly affected by the biocompatibility of these materials, unless we change to a material that supports microbial growth.

Cleanliness

Cleaning is commonly understood as the removal of soils such as dust, oils, dirt, etc. by chemical or mechanical means. Some of these soils will be visible, while others require a microscope or chemical processing to be seen.

Ideally, you won’t have to clean your medical device or its parts, but sometimes it’s necessary. Cleanliness requirements are entirely based on device application—devices that are unlikely to cause infection or interact with internal tissues probably don’t need to be very clean. Devices that interact with areas of the body that can’t readily process soils should be as free of contaminants as possible. 

Intuitively, one would think cleaning also removes microbes (reduce bioburden), toxins, and unwanted chemicals, but sometimes the process actually introduces more of these elements. Therefore, it’s important to solidify any cleaning processes before final bacterial endotoxin, sterility, and biocompatibility testing.

Gluco-Quick injects glucose into the bloodstream, so we need to be sure the materials are clean enough to meet our needs (e.g., comply with a standard like USP 788). We’ll start by testing all the components and the final product to see if they are clean enough, and if not, we’ll want to be sure to define our cleaning process ahead of endotoxin and biocompatibility testing as those factors will likely be affected (both in magnitude and variability) by cleaning.

Sterile Barrier

Many sterile barriers are embodied as sealed pouches or trays, but they can also be in the form of caps or other designs. As the name implies, sterile barriers keep microbes out, which means they also must keep out soils that could carry microbes, such as dust and liquids. In many cases, they also allow the air to pass in and out to prevent pressure buildup and facilitate gas sterilization methods. Sterile barriers are an element of overall device packaging, but they are not equivalent.

Though this section may seem out-of-sync with the rest of this article, it’s actually closely related. Depending on the specifics, sterile barriers can impact bioburden, cleanliness, and bacterial endotoxins simply through the course of handling during packaging and the introduction of large surface areas that could be covered in various microscopic elements. While it’s possible that the barrier could also affect biocompatibility, this is generally not the case. Finally, sterile barrier integrity is closely intertwined with sterilization dose as seals tend to be degraded by more intense sterilization.

Our Gluco-Quick device is sealed in a formed plastic pouch with a Tyvek lid. We must ensure the packaging is well defined before final endotoxin, bioburden, sterilization validation, and cleanliness testing, as the packaging can affect those factors. In this case, our strategy is to cap off the device before packaging to limit the impact of packaging handling on endotoxin, bioburden, and cleanliness.

Accelerated Aging

Until time travel is invented, ASTM F1980 will commonly be followed to mimic aging using the Q10 theory. This theory states that every 10°C increase in temperature doubles the chemical reaction rate. Roughly speaking, a month of heat treatment can simulate about a year of aging to support getting a device on the market quickly.

Again, this section may seem out of place, but if you think of this process as “heat treating” instead of “accelerated aging”, it becomes obvious how it can affect the human compatibility of your system. For example, heat treating may directly affect the device’s material characteristics (including biocompatibility), or indirectly by damaging the sterile barrier. Heat can also evaporate residues, chemical cleaners, or melt lubricants applied during cleaning or assembly. Failure to consider the effects of heat treating on these vectors can mean invalid results and a huge waste of time and money.

Aging should not affect bioburden or bacterial endotoxin levels, unless the device supports microbial growth over time. Aging may affect biocompatibility, though not common, and will typically negatively impact sterile barrier seals.

For Gluco-Quick, let’s pretend we thought we knew what the heat-treating process would do to the device, but after a month of heat exposure it turned out that the glucose experienced a drop in viscosity and leaked out past the piston. To compensate, we decided to change from a rubber seal to a metal piston ring.

Unfortunately, we had to repeat all of our endotoxin, bioburden, sterility validation, biocompatibility, and cleanliness testing. At least the package worked…

Production Environment

The production environment, in the context of this article, includes the work surfaces, temperature, humidity, air quality, storage, lighting, pests, handling processes and people, testing, etc. that are present during device assembly. This includes production environments for all constituent parts and storage of finished goods. For best results, qualify, monitor, and control as many of these factors as possible.

While some are more impactful than others, all together the production environment can have a huge effect on bioburden, bacterial endotoxins, and cleanliness, and can have some effect on the sterile barrier due to (UV) radiation, temperature, and humidity.

After a monstrous delay in our schedule due to the change in seal, we finally tested Gluco-Quick with the new design. For better or for worse, while we were doing that, we found a whole slew of new customers who wanted to buy the product, and our sales projections mean we need to change to a bigger facility. Who wants to tell the CEO that we will need to repeat, at a minimum, bioburden, endotoxin, and cleanliness testing?

By the end, I hope you gain a better understanding of each of the processes, how they are linked (if at all), and the potential paradoxes involved in trying to test everything at once.

Final Plan

For Gluco-Quick, a plan like this would be reasonable in the order shown:

1. Test all the glucose-touching subcomponents for endotoxins and sterilization compatibility
2. Test biocompatibility and cleanliness.
3. Establish the production environment and packaging for bioburden testing and complete sterilization validation.
4. Circle back and test the finished product for endotoxins, biocompatibility and cleanliness, and hope we sufficiently derisked those things in steps 1 and 2.

The interrelations of human compatibility factors can result in a paradox of which tests must come first. While this non-linearity means it’s impossible to completely remove risk, always consider and define testing order when developing your verification plan to remove as much risk as possible.

Sometimes the chicken comes before the egg, sometimes the egg comes first, and sometimes it’s more of a chicken-egg-chicken again situation. It’s tough but not impossible—once you’ve got it all working, try not to mess with it!

MORE FROM THIS AUTHOR: Catheter Sterilization for Disposable Devices

Nigel Syrotuck is a StarFish Medical Project Engineer.