Industries / Health / Medical devices
Materials and sterilization
Choosing what a device is made of so the body tolerates it, and killing the microbes on it, before first use and between uses, without damaging it.
7
Milestones achieved, from 1952 to 2026.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1952
Artery tubes from cloth
The surgeon Arthur Voorhees saw that a smooth lining formed on sutures left in the bloodstream and reasoned that a fabric tube might grow the same lining. He sewed tubes of Vinyon-N, a vinyl plastic cloth, on his wife's sewing machine, reported working aortic grafts in dogs in 1952 and a first implant in a person in 1954. Nylon grafts that followed broke down in the body, and woven or knitted Dacron, a polyester, came to dominate.
LessonNylon failed because its chemical backbone is the same as that of proteins, which the body knows how to break down; as of 2023 no regulator had approved a synthetic artery graft 4 mm or less across, by Buddy Ratner's review.
Credit
Arthur Voorhees, with Arthur Blakemore
1962
PTFE hip sockets fail
From 1956 the British surgeon John Charnley lined artificial hip sockets with PTFE, a slippery plastic. In the body it wore away at up to 0.5 mm a month and its wear debris set off an intense tissue reaction, problems found only after implants in about 300 patients, and a glass-filled version that did well in laboratory tests also failed. One of his assistants tested ultra-high-molecular-weight polyethylene, a plastic then used in loom bearings, and Charnley implanted the first socket made of it in November 1962.
LessonLaboratory wear tests predicted success for PTFE, and the body showed otherwise.
Credit
John Charnley and his team
2015
Scopes too intricate to clean
A duodenoscope is a flexible camera tube passed through the mouth into the small intestine and reused after cleaning. On 19 February 2015 the FDA warned that its complex design may impede reprocessing (cleaning and disinfection between patients), after infections linked to reprocessed scopes. Studies the FDA ordered found more contamination after reprocessing than expected, and in August 2019 the FDA recommended a move to scopes with disposable end caps or fully disposable scopes, as Olympus, Fujifilm and Pentax withdrew their fixed end-cap models.
LessonThe fix the FDA backed was a design that is easier to clean or is thrown away, not stricter cleaning alone.
Credit
US FDA; Olympus, Fujifilm and Pentax
2017
A dissolving stent withdrawn
Abbott's Absorb was a scaffold for holding open narrowed heart arteries that dissolves over about three years instead of staying as metal. Trials found more clots in the device and more heart attacks in the treated artery than with metal drug-coated stents, and Abbott withdrew it from the market in September 2017. Pooled five-year results from all five Absorb trials, in 5,998 patients, put the failure rate at the treated site at 15.9% against 13.1% for metal stents, with the extra risk confined to the first three years.
LessonA material that disappears has to stay safe while it disappears, and the risk here came in the years the scaffold was still being absorbed.
Credit
Abbott; the ABSORB trial investigators
2019
A sterilizer sealed shut
Ethylene oxide (EtO) is a gas that kills microbes without heat, and by the FDA's count it sterilizes about 50% of US devices that need sterilizing, more than 20 billion a year. After US EPA air monitoring in 2018 found high EtO levels near a Sterigenics plant in Willowbrook, Illinois, the State of Illinois sealed the plant on 15 February 2019, and the company announced its closure that September. The FDA issued statements on preventing device shortages and began pushing makers toward other methods.
LessonWhen a few large plants sterilize a large share of devices, the loss of one puts supply at risk.
Credit
State of Illinois and US EPA; FDA
2024
Hydrogen peroxide joins the list
On 8 January 2024 the FDA added vaporized hydrogen peroxide, a gas that sterilizes at low temperature, to its Established Category A sterilization methods, beside moist heat, dry heat, EtO and radiation, after recognizing the standard ISO 22441:2022. The FDA said the change would ease wider use of the method as part of its push to reduce EtO use.
Credit
FDA; the International Organization for Standardization (ISO)
2026
Biocompatibility judged by risk
Biocompatibility is evidence that a device's materials do not harm the body. On 25 May 2026 the FDA recognized, in part, the 2025 edition of ISO 10993-1, which starts from the device's chemistry and how the body is exposed to it and places the evaluation inside the maker's overall risk management, in place of a fixed table of tests. The FDA left out a clause on estimating biological risk that it said conflicts with ISO 14971, and accepts the 2018 edition until 1 July 2029.
Credit
ISO; FDA
Watch for
EPA moves to repeal limits
In April 2024 the EPA issued a rule to cut EtO emissions from commercial sterilizers by more than 90%, giving the 28 largest users (over 60 tons a year) two years to comply. In March 2026 it proposed repealing the standards that came from that rule's health-risk review while keeping those from its review of control technology, saying EtO sterilizes 50% of US devices, about 20 billion a year, with "no viable alternative on the market", and asking whether its 2016 cancer-risk value for EtO is still right. The comment period closed on 15 May 2026.
Achieved whena final rule is published in the Federal Register; none had been by 17 November 2026.
Claimed by
US Environmental Protection Agency, March 2026
Open
01
An open sterilization damage table
A public, machine-readable table of how device plastics, rubbers and adhesives change after each sterilization method (steam, EtO, gamma rays, electron beam, vaporized hydrogen peroxide): material grade, method, dose or cycle, property measured, change in percent, and the paper, table and page it came from.
The briefCredit · unclaimed02
Printable cleaning test pieces
A public set of printable or machinable test pieces, each carrying one hard-to-clean feature of reusable devices (a narrow channel, a hinge gap, a screw thread, a bearing pocket, two surfaces pressed together), with a fixed soiling and cleaning procedure and an open table of how much residue each feature keeps after each cleaning method. A designer could then look up a feature's cleaning risk before the design is frozen.
The briefCredit · unclaimed
Open 01 · the brief
An open sterilization damage table
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public, machine-readable table of how device plastics, rubbers and adhesives change after each sterilization method (steam, EtO, gamma rays, electron beam, vaporized hydrogen peroxide): material grade, method, dose or cycle, property measured, change in percent, and the paper, table and page it came from.
Why it is still open
Every switch away from EtO needs evidence that the device's materials survive the new method. The reference most often quoted, AAMI's technical report TIR17, ranks polymers by their resistance to radiation rather than giving measured changes, and resin makers publish guides for their own products. The nearest compilation is a paid databook, Liesl Massey's The Effect of Sterilization Methods on Plastics and Elastomers, last revised in 2004, with data on 43 materials in graphs and tables. Newer measured numbers sit in single papers, such as a 2016 comparison of EtO, hydrogen peroxide, gamma and electron beam on bioresorbable polymers. Searches of the web and GitHub found no open dataset of property changes across methods.
shop.elsevier.commedicaldesignbriefs.compmc.ncbi.nlm.nih.gov
Built with what exists
the Europe PMC and OpenAlex APIs to find papers; Claude or Gemini to pull tables from PDFs into a fixed format with page citations, an advance that turns weeks of hand copying into checking; two people to check samples; DuckDB and a public repository such as GitHub or Zenodo; for spot checks, standard plastic test bars sent through a contract sterilizer and pulled on a university tensile tester.
First step
Pull 50 open-access papers on sterilizing common device plastics from Europe PMC, copy 100 rows by hand, and compare a model's extraction with yours.
Done looks like
At least 1,000 rows covering 30 or more materials and five methods, each traced to its source page. A second person checks 200 random rows and agrees with at least 95%. Ten rows are re-measured on test bars, and the measured change falls inside the table's range for at least eight.
Open 02 · the brief
Printable cleaning test pieces
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public set of printable or machinable test pieces, each carrying one hard-to-clean feature of reusable devices (a narrow channel, a hinge gap, a screw thread, a bearing pocket, two surfaces pressed together), with a fixed soiling and cleaning procedure and an open table of how much residue each feature keeps after each cleaning method. A designer could then look up a feature's cleaning risk before the design is frozen.
Why it is still open
Kremer, McDonnell and colleagues, with Johnson & Johnson staff, showed that cleaning results depend on a device's features, testing features with 56,000 flushes in 2023 and finding in 2025 that hidden features such as ball bearings, leaf springs, threads and mated surfaces were hardest to clean. They proposed a database of design features; their data sit inside the articles, and searches of the web, Europe PMC and GitHub found no shared design files for test pieces or open results table. Test devices for the cleaning process exist, such as Aseptium's patented capsule that holds a soiled tag inside a narrow channel (US patent 11,957,809, granted April 2024), but they check whether a cleaning cycle works, not which design feature holds soil. In 2019 the FDA said ATP test strips, a common quick check of cleanliness, had not been evaluated for duodenoscope cleaning.
patents.google.compmc.ncbi.nlm.nih.govpmc.ncbi.nlm.nih.govfda.gov
Built with what exists
a resin 3D printer, or a print service for stainless steel parts; PTFE tubing for channels; a test soil made to ISO 15883-5:2021 Annex A (coagulated sheep blood, no microbes), as Beaugelin and Pineau used in a 2026 brush study; a protein assay kit and a plate reader; a syringe pump for flushing; a motorized brush stage, or a low-cost robot arm such as Hugging Face's SO-101 taught the brushing by demonstration so every pass is the same (see the Note "Five demonstrations per task" and the Note "No download").
First step
Soil PTFE tubes of four bore sizes, brush them as a scope maker instructs and with one pass, and check your protein numbers against Beaugelin and Pineau's.
Done looks like
Design files for 20 test pieces covering five kinds of feature, each soiled and cleaned three ways (brushing as instructed, a single brush pass, soaking with ultrasound) five times, with residual protein in micrograms per square centimeter published for every run. A second lab prints the files and gets the same order for the five hardest pieces.