Biocompatibility testing is a series of laboratory and sometimes animal-based evaluations designed to determine whether a medical device material causes harmful biological reactions when it contacts the human body. Every medical device that touches skin, tissue, or blood, from a simple adhesive bandage to an implanted heart valve, must go through some level of this testing before it can reach the market. The international standard that governs the process is ISO 10993, a multi-part framework that the FDA uses as its foundation for evaluating device safety in the United States.
How Devices Are Categorized
Not every device needs the same tests. The starting point is classifying the device by two factors: what part of the body it contacts and how long it stays there. The nature of body contact falls into three broad groups. Surface devices touch intact skin, mucous membranes, or compromised skin. External communicating devices have pathways into the body, contacting tissue, bone, or circulating blood without being fully implanted. Implant devices are placed inside the body in direct contact with tissue, bone, or blood.
Contact duration adds a second layer. A device worn for less than 24 hours is classified as limited exposure. Anything from 24 hours to 30 days counts as prolonged. Beyond 30 days is long-term. A device that sits on intact skin for a few hours will need far fewer tests than a permanent implant bathed in blood. The combination of contact type and duration produces a matrix that maps directly to which biological endpoints need evaluation.
Cytotoxicity: The First Screen
Almost every biocompatibility evaluation starts with cytotoxicity testing, a cell-based lab test that checks whether the device material kills or damages living cells. It is the most common in vitro (meaning “in the lab,” not in a living animal) biocompatibility test, and it serves as an early warning system. If a material fails here, more complex testing is unlikely to proceed.
The standard method involves extracting chemical compounds from the device material into a liquid, then exposing mouse fibroblast cells (a common lab cell line called L929) to that extract. Technicians examine the cells under a microscope for changes in shape, structure, and overall survival. Each sample gets a score from 0 to 4 based on how severe the damage is. A score of 2 or less is considered passing.
A second approach, more common in European regulatory submissions, measures cell metabolic activity using a colorimetric dye. Healthy cells convert the dye into a colored compound, and the intensity of that color reflects how many cells are alive and functioning. A result showing 70% or greater cell viability passes this version of the test. Both methods aim to answer the same question: does this material poison cells on contact?
Sensitization and Irritation
Beyond direct cell toxicity, regulators want to know whether a device will trigger an allergic reaction or irritate the tissue it contacts. Sensitization testing evaluates the potential for a delayed immune response, the kind that builds over repeated exposures and eventually causes a reaction like contact dermatitis. Irritation testing looks for more immediate local damage: redness, swelling, or tissue destruction at the contact site.
These assessments have traditionally relied on animal models, but the field is shifting. ISO 10993-10, the part of the standard covering irritation and sensitization, now includes provisions for computer-based modeling and lab-based alternatives for certain types of skin exposure. The goal is to reduce animal testing where validated alternatives exist, while still ensuring that devices contacting mucous membranes or internal tissues are rigorously evaluated.
Systemic Toxicity
Some devices release chemicals that can travel throughout the body rather than causing only local effects. Systemic toxicity testing checks whether those substances harm organs or systems far from the contact site. The testing is broken into timeframes that mirror how long a patient would be exposed.
Acute systemic toxicity studies involve a single high dose of the device extract, followed by close observation for up to 14 days. Technicians look for signs of illness, organ damage, and delayed effects. Beyond acute studies, subacute and subchronic protocols extend exposure to match the device’s intended use period, monitoring for toxic effects that only emerge over time. At the end of each study, tissues are examined microscopically to identify which organs, if any, were affected. These tests are particularly important for implants and devices that contact circulating blood, where chemical leaching is a real concern.
Blood Compatibility
Any device that contacts blood, whether directly or through tubing and catheters, needs hemocompatibility testing. Blood is a complex living tissue, and foreign materials can trigger several distinct problems. The evaluation looks at multiple pathways: whether the material promotes blood clot formation, whether it activates the clotting cascade, whether it causes platelets to clump together, and whether it triggers the complement system (part of the immune response that can cause inflammation).
These tests use human blood samples circulated over or through the device material under controlled conditions. Researchers measure activation markers before and after contact. A material might perform well on clotting but poorly on complement activation, so each pathway is assessed independently. This makes hemocompatibility one of the more complex and resource-intensive categories of biocompatibility evaluation.
Chemical Characterization as an Alternative
One of the most significant shifts in biocompatibility testing over the past decade is the growing role of chemical analysis. Rather than relying solely on biological tests to detect harmful effects, manufacturers can now identify exactly which chemicals leach out of a device and assess their toxicity using existing safety data. The FDA explicitly recognizes this approach as an alternative to biological testing for several endpoints, including acute, subacute, subchronic, and chronic systemic toxicity, as well as genotoxicity, carcinogenicity, and reproductive toxicity.
The process works in two stages. First, extractable and leachable studies identify chemicals that migrate out of the device under aggressive lab conditions (extractables) and under conditions that simulate real clinical use (leachables). Analytical techniques detect both known target compounds and unexpected substances. Second, a toxicological risk assessment compares the amount of each substance a patient would be exposed to against established safety thresholds. If all identified chemicals fall below levels of concern, the device can clear certain biocompatibility endpoints without additional animal testing.
This approach is governed by ISO 10993-18 for the chemical analysis and ISO 10993-17 for the risk assessment. The FDA recognized the updated second edition of ISO 10993-17, published in September 2023, and will accept conformity declarations to the older first edition only until December 2026. After that, all submissions must align with the new version, which refines how allowable exposure limits are calculated for device constituents.
Sample Preparation Matters
A detail that can make or break test results is how the device sample is prepared for extraction. Standards specify precise ratios between the device’s surface area and the volume of extraction liquid. A small nitinol staple with a surface area of about 19 square millimeters, for instance, would be extracted at a ratio of 6 square centimeters per milliliter of solvent. Getting this ratio wrong can either concentrate harmful chemicals beyond realistic exposure levels or dilute them so much that real risks go undetected. The extraction conditions, including temperature, solvent type, and duration, are all standardized to ensure results are reproducible and comparable across laboratories.
What the FDA Requires
In the United States, biocompatibility data is required for virtually every type of device submission: 510(k) premarket notifications, premarket approval applications, investigational device exemptions, humanitarian device exemptions, and De Novo requests. The FDA uses ISO 10993-1 as its reference framework but publishes its own guidance document that clarifies where agency expectations differ from or add to the international standard. Some parts of ISO 10993-17, for example, received only partial FDA recognition, with specific tables and calculation examples excluded.
The scope of testing required depends entirely on the device classification matrix. A manufacturer can consult the FDA informally by email or through a formal pre-submission (Q-Sub) process to confirm which endpoints apply to their specific device before committing to expensive testing programs. This step can save months of work and significant cost, since a full biocompatibility testing package for a long-term implant can involve a dozen or more individual studies spanning cytotoxicity, sensitization, irritation, systemic toxicity, hemocompatibility, genotoxicity, and implantation effects.