ISO 10993 is the standard family that governs the biological evaluation of medical devices — the process that establishes whether the materials touching a patient are safe. What you have to test is determined by two things: the nature of the body contact (skin surface, breached surface, blood path, implant) and how long that contact lasts (under 24 hours, up to 30 days, or longer). Those two axes produce a test matrix, and the difference between the mildest and the most demanding corner of that matrix is a difference of one or two orders of magnitude in cost and many months of schedule.

This article is educational only. Projects House is an engineering firm, not a regulatory consultancy or testing laboratory; a qualified toxicologist or regulatory professional should determine the biological evaluation plan for your specific device.

Why This Is Its Own Discipline

Engineers are used to asking whether a material is strong enough, moldable, and affordable. Biocompatibility asks a completely different question: what does this material release into the body, and what does the body do in response? A polymer that is mechanically ideal may carry a plasticizer, colorant, mold release agent, or residual catalyst that fails a cytotoxicity screen.

Crucially, the evaluation applies to the finished device in its final, sterilized state — not to raw material data sheets. The same base resin, processed differently or sterilized differently, can produce different results. A supplier's claim that a resin is "medical grade" is a starting point, not evidence.

The modern approach is a biological evaluation plan built on a risk assessment: you identify what a patient could be exposed to, gather existing data where it is credible, and test only the gaps. That framing sits directly inside the process described in ISO 14971 risk management.

What Determines the Level of Testing

Contact category, in ascending demand:

  • Surface contact — intact skin, mucous membranes, or breached and compromised surfaces. A wearable strap sits at the mild end; a wound dressing does not.
  • External communicating — contact with a blood path, tissue, or bone via an indirect route. Tubing sets, catheters, and dialysis components fall here.
  • Implant — devices placed in tissue, bone, or the bloodstream. The most demanding category by a wide margin.

Contact duration: limited (under 24 hours), prolonged (24 hours to 30 days), or long-term (over 30 days). Duration is cumulative and counts repeated exposures, so a device used for twenty minutes a day every day does not qualify as limited contact.

A skin-contact device used briefly may need only a small set of screening endpoints. A long-term blood-contacting implant can require the full battery including chronic systemic toxicity, genotoxicity, implantation studies, and hemocompatibility. Getting the category wrong in your planning is how a device program discovers a year of unbudgeted work.

What Actually Gets Tested

The commonly required endpoints, roughly in order of how often they appear:

  • Cytotoxicity — whether extracts of the device harm cultured cells. Almost universally required, relatively fast and inexpensive, and the most common first failure.
  • Sensitization — potential to cause an allergic response on repeated exposure.
  • Irritation or intracutaneous reactivity — local tissue response.
  • Acute systemic toxicity — effects of a short, high exposure.
  • Material-mediated pyrogenicity — fever response attributable to the material.
  • Hemocompatibility — hemolysis, coagulation, and platelet effects, for anything in the blood path.
  • Genotoxicity, subchronic and chronic toxicity, carcinogenicity, and implantation — reserved for longer-duration and implanted devices, and by far the most expensive and slowest endpoints.

Underlying most of this is chemical characterization: extractables and leachables analysis identifying what actually comes off the device under exaggerated conditions, followed by a toxicological risk assessment of each identified compound. Regulators increasingly expect this chemistry-led route, and it can replace some biological testing when the analysis is thorough.

Materials With a Track Record

Choosing from the set of materials with extensive device history is the cheapest risk reduction available:

  • Polymers: medical-grade silicone, polycarbonate, polypropylene, PEEK, PTFE and related fluoropolymers, polyurethanes formulated for device use, and certain polyesters.
  • Metals: implant-grade stainless steels, titanium and its alloys, cobalt-chromium, and nitinol for specific applications.
  • Adhesives, inks, and coatings: the most frequently overlooked category and a very common source of failures, because they are chosen late by whoever is solving an assembly problem.

The specification that matters is not the resin family but the exact grade, the exact colorant, and the exact processing conditions. Ask suppliers for a master file reference or prior biological data, and pin the material down in your documentation to a level that prevents a molder from substituting an equivalent grade. General material selection tradeoffs are covered in how to choose materials for a new product and how to choose plastic for your product; for devices, biocompatibility overrides most of the usual criteria.

Sterilization Is the Decision That Drives Everything

Sterilization method and material choice cannot be decided independently, and choosing sterilization late is one of the most expensive sequencing errors in device development.

  • Ethylene oxide is gentle on most polymers but leaves residuals that must be shown to be within limits, and requires packaging that allows gas exchange.
  • Gamma and electron beam radiation are fast and residue-free but degrade several common polymers, causing embrittlement, discoloration, or reduced molecular weight — sometimes progressively over shelf life.
  • Steam autoclaving is cheap and clean but rules out any material or adhesive that cannot survive the temperature and moisture, and it constrains electronics and batteries severely.

Because the biological evaluation applies to the sterilized device, changing sterilization method after testing generally means repeating the testing. Lock the method during early design, and validate sterility and shelf life on the final configuration.

How to Reduce the Testing Burden

  • Reduce contact. If a coating, cover, or barrier layer means only one well-characterized material touches the patient, the entire evaluation shrinks to that material.
  • Use materials with existing device history and available data. Prior evidence on an identical grade, processed and sterilized the same way, can substitute for new testing.
  • Lead with chemical characterization. A rigorous extractables study plus toxicological assessment can close endpoints that would otherwise require biological testing.
  • Reduce the number of distinct patient-contacting materials. Every additional material is another evaluation.
  • Test the final configuration once. Testing early prototypes and then changing a colorant, an adhesive, or a mold release means paying twice.
  • Write a biological evaluation plan before testing. It is the document that justifies which endpoints you are not running, and it is what a reviewer reads.

Cost ranges honestly: a limited-contact device with well-known materials may need only a few thousand to low tens of thousands of dollars of testing. A long-term implant with novel materials runs into the hundreds of thousands and takes many months, because some endpoints are inherently long studies that cannot be compressed. Those figures belong in your program budget from the beginning — see medical device development cost and, for how the category is determined, FDA medical device classes.

What This Means for the Design Itself

Biocompatibility constrains engineering in ways that surprise teams: colorants may be off the table, common adhesives replaced by mechanical joints or welding, mold releases prohibited, and surface finishes limited to those achievable without questionable secondary processes. Design changes made for cost or manufacturability late in a program can invalidate completed testing, which is why device programs freeze patient-contact materials earlier than consumer product programs do. Building and evaluating early units in the intended final materials, as described in medical device prototyping, is what keeps this manageable — and the whole chain of decisions is recorded through the process in FDA design controls.

Projects House designs patient-contact devices with material, adhesive, and sterilization decisions made together and early, so the biological evaluation is planned rather than discovered. If you are developing a device that touches a patient, tell us about it through our contact form and we will help you scope the material and testing path.