health · manufacturing · family: making one is easy. making a million is the problem
printed one byone, regulated by none
3D-printed medical devices are made one-at-a-time in hospitals — but regulations assume factory assembly lines
Problem statement
Additive manufacturing (3D printing) is increasingly used to produce patient-specific medical devices — anatomical models, surgical guides, prosthetics, implants — and emerging bioprinting technologies aim to fabricate living tissue constructs. But the FDA's regulatory framework was designed for mass-manufactured devices produced under controlled factory conditions, not for one-off devices fabricated at the point of care in hospital 3D printing labs. It is unclear whether a hospital printing a patient-specific surgical guide is a "manufacturer" subject to FDA oversight, what quality management system requirements apply, and how postmarket surveillance works for one-of-a-kind devices. For bioprinted constructs incorporating living cells, the FDA's biologics center had neither approved any 3D-printed biological product nor published specific guidance on bioprinting as of Pew's 2022 brief, and the 2017 guidance leaves biological, cellular, and tissue-based products out, saying they may need different regulatory pathways.
Why this matters
The use of 3D printing within hospitals steadily expanded over the decade before Pew's 2022 brief, producing patient-specific anatomical models, surgical guides, prosthetics, and implants, and one concern from Pew's interviews is that such devices may be manufactured at hundreds of sites by entities, such as hospitals, that the FDA does not typically regulate. Point-of-care manufacturing offers on-demand surgical guides and anatomically matched implants — real benefits in smaller-scale studies, including shorter operating times and improved patient outcomes — but Pew warned that until the FDA issues clear guidance, health care facilities will keep producing devices that may not meet its standards for safety and effectiveness. Bioprinting, still in the initial stages of research and development, may one day enable the creation of personalized organs and other body parts, and Pew notes that future guidance for point-of-care devices could have implications for it.
What’s been tried and why it hasn’t worked
The FDA issued guidance on "Technical Considerations for Additive Manufactured Medical Devices" in 2017, but it says only that point-of-care manufacturing may raise additional technical considerations and does not address them. In a December 2021 discussion paper the FDA outlined three scenarios for point-of-care printing — a health care facility using a production system supplied by a traditional manufacturer, a traditional manufacturer co-located at or near the facility, and a facility taking on all of a traditional manufacturer's responsibilities — but Pew found the framework short on detail, and as of its 2022 brief there was no draft or formal guidance. The custom device exemption is, in Pew's words, an extremely narrow category (no more than five units of a device per year, among other requirements), and the FDA has emphasized that a personalized 3D-printed device qualifies only if it meets all its criteria. ASTM's F3659, "Standard Guide for Bioinks Used in Bioprinting" (ASTM page dated April 2024), is a guide to bioinks for tissue-engineered medical products, organized as pre-printing, printing, and post-printing considerations; extrusion bioprinting is its primary focus, though other modalities are also addressed. Pew's interviews with 17 experts from medicine and product manufacturing, conducted February to April 2021, found the FDA's draft framework ambiguous on how device risk will be defined and on which regulations apply to whom. In the absence of FDA oversight, many of the experts Pew interviewed agreed that quality assurance for the lowest-risk products would fall to the health care facilities themselves, which would need to develop and implement their own processes.
What would unlock progress
A regulatory framework that distinguishes between factory-based and point-of-care additive manufacturing — with clear registration, quality management, and validation requirements for hospital-based 3D printing — would resolve the current jurisdictional ambiguity. For bioprinted products, a classification pathway that accounts for the device-biologic combination (the 2017 guidance sends questions on biologics, cells, and tissue products to CBER, and questions of jurisdiction over a combination product to the Office of Combination Products) would settle a question the 2017 guidance leaves open. Non-destructive testing methods capable of validating patient-specific geometries without destroying the actual device would address a core quality assurance challenge; the 2017 guidance lists non-destructive evaluation techniques (ultrasound, CT, and x-ray among them) for verifying the geometry of individual devices, and notes that some are not suitable for some materials, designs, or intended uses.
Entry points for student teams
A student team could survey 3D printing labs across multiple hospitals to document current quality management practices, identify common failure modes, and propose a minimum-viable quality standard for point-of-care manufacturing. Alternatively, a team could develop a non-destructive testing protocol for a specific class of 3D-printed device (e.g., surgical guides), comparing printing parameters, material properties, and dimensional accuracy across different printer platforms. A design-focused team could prototype a quality assurance workflow — from digital file to finished device — that a hospital 3D printing lab could adopt.
Genes
Structural cousins: stuck the same way, in other fields
Sources
Pew Charitable Trusts, "FDA's Regulatory Framework for 3D Printing of Medical Devices at the Point of Care Needs More Clarity" (2022), FDA, "Technical Considerations for Additive Manufactured Medical Devices" (2017), accessed 2026-02-19 go to source ↗
Working notes
Notes made while writing and checking this brief, shown as written. They use working shorthand.
Primary sources are Pew Charitable Trusts analysis and FDA guidance documents. The regulatory-mismatch failure mode is dominant — the regulatory framework was built for a manufacturing paradigm (batch production) that point-of-care 3D printing fundamentally violates. The lab-to-field-gap captures the disconnect between factory-validated processes and hospital-floor execution. Manufacturing is added as a secondary domain because the core problem is about manufacturing quality control, not just clinical outcomes. The not-attempted tag reflects that, as of Pew's 2022 brief, the FDA's biologics center had published no specific guidance on bioprinting.
Reconciliation 2026-09-30: Whole write-up read against the sources it names. Read in full, fetched 2026-09-30: the Pew issue brief of July 27, 2022 (the Pew site blocks automated access, so the Internet Archive copy of the same page was read: https://web.archive.org/web/20250813221951/https://www.pew.org/en/research-and-analysis/issue-briefs/2022/07/fdas-regulatory-framework-for-3d-printing-of-medical-devices-needs-more-clarity); the FDA guidance of December 5, 2017 (https://www.fda.gov/media/97633/download, 32 pages). Read in part: the ASTM catalogue pages for the standard named in the text (https://store.astm.org/f3559-24.html and https://store.astm.org/f3659-24.html), which show the scope and contents list but not the standard itself. Changes: (1) The Source line gave the Pew title without its last three words; it now carries the full title from the page, "... at the Point of Care Needs More Clarity." (2) The device examples "cranial plates, spinal cages, dental prosthetics" are not in either source; the list now gives Pew's examples (anatomical models, surgical guides, prosthetics, implants). (3) "No FDA-approved product exists and no regulatory pathway adequately addresses the unique risks of combining device and biologic properties" went further than the sources. Pew says the FDA's biologics center had not approved any 3D-printed biological product and had not published specific guidance on bioprinting; the 2017 guidance says it does not cover biological, cellular or tissue-based products and that these may need different regulatory pathways. The sentence now says that, dated to 2022. (4) "Over 100 U.S. hospitals now operate 3D printing labs" is not in either source. Pew says hospital use steadily expanded over the decade before its brief and that devices may be manufactured at hundreds of sites; the sentence now says that, dated to 2022. (5) "The global medical 3D printing market is projected to exceed $5 billion by 2028" is not in either source and was removed. (6) "Same-day" surgical guides is not in the sources; Pew says "on-demand." "Uncontrolled quality means defective devices may reach patients" was stronger than Pew, which says facilities will continue to produce products that may not meet FDA's standards until guidance is issued; the sentence now follows Pew. (7) That bioprinting could "address organ transplant shortages" and that "regulatory uncertainty is cited as a primary barrier to clinical translation" are not in the sources. Pew says bioprinting is in the initial stages of research and development, may one day enable personalized organs, and could be affected by future point-of-care guidance; the sentence now says that. (8) The text said the 2017 guidance "was written for factory-based additive manufacturing." The guidance does not say that; it says point-of-care manufacturing "may raise additional technical considerations that are not addressed in this document." The sentence now says that. (9) The text said the FDA "has suggested that point-of-care devices might be regulated as custom devices ... or as manufacturer-specific cleared devices." The sources say the opposite about custom devices: both state that a patient-matched device is not a custom device unless it meets all the criteria of section 520(b), and Pew calls the exemption "an extremely narrow category" (no more than five units a year, among other requirements). What the FDA did propose, per Pew, is a framework of scenarios, cut from five to three in a December 2021 discussion paper, with no draft or formal guidance at the time Pew wrote. The passage now describes the three scenarios and, in Pew's words, the extremely narrow custom-device exemption. (10) The standard was named as ASTM F3559. The ASTM page for F3559 is a guide to amusement-ride standards. The bioink standard is F3659, "Standard Guide for Bioinks Used in Bioprinting," page dated April 16, 2024. Its scope says extrusion bioprinting is "the primary focus" but "other bioprinting modalities are also addressed," so "covers only extrusion bioprinting" was wrong, and the page does not call it the first such standard or say it is mainly about terminology. The sentence now gives the designation, title, structure and scope from the ASTM page; the page's boilerplate line that the user must determine the applicability of regulatory limitations was not kept, since it is not a finding about this standard. (11) "A 2025 industry survey found that manufacturers consider FDA additive manufacturing guidance ambiguous" names no source and is not in the sources read. It was replaced with what Pew reports from its own interviews: 17 experts, February to April 2021, who found the draft framework ambiguous. (12) "Hospital-specific institutional protocols that vary widely, with no standardized validation methodology" is not in the sources. Pew reports that many experts agreed health care facilities would themselves be responsible for quality assurance on the lowest-risk products; the sentence now says that. (13) "Currently falling between CDRH and CBER jurisdictions" is not stated in the sources; the parenthesis now gives what the 2017 guidance says (biologics questions go to CBER, combination-product jurisdiction to the Office of Combination Products), and "would unblock clinical translation" became "would settle a question the 2017 guidance leaves open," since the guidance says it does not address such products and that they may need different regulatory pathways. (14) Non-destructive testing "would solve the fundamental quality assurance challenge" was softened; the 2017 guidance lists non-destructive techniques for checking the geometry of individual devices and says some do not suit some materials, designs or uses. (15) In the working notes above, "no pathway exists" now reads as what Pew says (no specific guidance on bioprinting as of 2022), and the sentence on international harmonization (EU MDR, PMDA) was removed because neither source discusses other countries. Confirmed in the sources and left as written: the 2017 date and title of the FDA guidance; that point-of-care products do not fit neatly into the FDA's standard oversight system, which Pew contrasts with large-scale plants producing high volumes of identical products; that it is unclear which regulations apply to a health care facility and when it counts as a manufacturer; the open questions about quality systems and postmarket surveillance; Pew's call for facility registration. Not changed but worth a second look: the display title says "Regulated by None," while Pew says FDA requirements apply to all medical devices however they are made, and that what is missing is clear guidance on how. A second reader went over these repairs on the same day, opened the same sources independently, and tightened four sentences: the hospital-expansion sentence is now dated to Pew's brief rather than to the present; the custom-device sentence attributes "extremely narrow" to Pew and drops an editorial framing; the ASTM sentence keeps the page's structure and scope and drops the boilerplate clause; the bioprinting-pathway sentence in What Would Unlock Progress no longer says a pathway "would remove that uncertainty," which named no uncertainty the parenthesis had stated, and instead says it would settle a question the 2017 guidance leaves open. Could not be checked: anything after July 2022. Both sources predate the write-up by years, and whether the FDA has since issued point-of-care guidance was not looked up, because no source named here covers it. Genome tags untouched.