EU MedTech startups crossing the Atlantic make the same seven mistakes with remarkable consistency: assuming a CE mark grants US market access, choosing a predicate device casually, translating the IFU instead of rewriting it, treating the US Agent as paperwork, assuming MDR clinical evidence is enough for the FDA, skipping a real US distribution partner, and ignoring the FDA's quality system expectations under the QSR/QMSR framework. Each one is avoidable. Each one is expensive when it happens.

By Tibor Zechmeister and Felix Lenhard. Last updated 10 April 2026.

TL;DR

  • A CE mark under Regulation (EU) 2017/745 Article 20 is a European conformity declaration. It has no legal standing with the FDA.
  • Predicate device selection is the single highest-leverage decision in a 510(k) submission, and the one EU founders most often get wrong on the first attempt.
  • The US Instructions for Use is not a translation of the EU IFU. It is a separate labelling artifact with its own content and format expectations.
  • The US Agent designation is a legal requirement for foreign manufacturers, not an administrative courtesy. The choice of US Agent has strategic consequences.
  • Clinical evidence prepared for MDR Article 61 and Annex XIV does not automatically satisfy FDA expectations on trial design, endpoints, or statistical analysis.
  • Selling into the US without a real distribution partner or commercial footprint is a failure mode that has nothing to do with regulation and everything to do with revenue.
  • The FDA's quality system framework (historically 21 CFR Part 820, now aligning with ISO 13485 under the QMSR initiative) is related to but not identical with the EU QMS expectations.

Why this post exists

A German MedTech company Tibor worked with a few years ago had built a clever device, secured CE marking, and walked into an investor meeting with an expansion deck that listed a dozen countries. The revenue model assumed parallel EU and US entry in the same year. The investor asked a single question: how does your CE mark help you in the US? The founder answered that CE was international. The investor, who had worked in US regulatory affairs, pushed back. The meeting ended without a term sheet. The company spent the following months rewriting the business plan, revising timelines, and cutting near-term revenue projections because the US market would not open on the schedule the deck had promised.

That company is not unusual. It is one case in a long series of EU founders who assumed that what worked in Europe would travel. This post collects the seven mistakes Tibor sees most often in that first meeting, written so the next founder can skip them.

A framing note before the list: this post stays at general framing on the FDA side. For any specific US pathway question on your device, work with a US regulatory specialist who practices inside the FDA system daily. Tibor's authority is the EU MDR and the Notified Body framework. The US specialists are the execution partners. This post is the orientation.

Mistake 1 — Assuming a CE mark means US market access

The CE mark declares that a device complies with Regulation (EU) 2017/745 and can be lawfully placed on the market in the EU and EEA. That is all it does. It does not satisfy the FDA. It does not satisfy Health Canada, the UK MHRA, Japan's PMDA, or any other national regulator. Each jurisdiction has its own framework, its own classification rules, and its own premarket process.

For the US specifically, CE has no legal role. An EU manufacturer selling a CE-marked device into the United States without an FDA clearance or approval is non-compliant with US law, regardless of how complete the CE technical documentation is. The underlying technical work done for CE — bench testing, biocompatibility, electrical safety, risk management, clinical data — can often be reused as supporting evidence inside an FDA submission, but the submission itself is a separate regulatory event that must meet FDA rules.

The fix is to treat the US as a parallel track with its own plan, its own budget, and its own timeline from the moment the US enters the business plan. Not after CE. From day one of the strategic decision.

Mistake 2 — Choosing a predicate device casually

For most Class II devices, the US pathway is a 510(k) premarket notification, and the heart of a 510(k) is the predicate comparison. The manufacturer identifies an existing legally marketed device and demonstrates that the new device is substantially equivalent to that predicate in terms of intended use and technological characteristics, or that any differences do not raise different questions of safety and effectiveness.

The mistake EU founders make is picking a predicate the way you pick a comparable in a pitch deck: by similarity of appearance, by shared buzzwords, by a quick search of the FDA database. That approach produces a 510(k) that fails on first review because the chosen predicate is either inappropriate for the intended use, outdated, withdrawn, or carries technological differences that require additional performance data the applicant did not anticipate.

Predicate selection is the highest-leverage decision in a 510(k) and the one where a US regulatory specialist earns their fee several times over in a single structured meeting. Done right, it locks the entire submission strategy. Done wrong, it forces a rework that can cost months.

Mistake 3 — Translating the IFU instead of writing a US version

The EU IFU is written against MDR Annex I Chapter III format, uses multi-language translations, and carries the CE mark and UDI in the EUDAMED format. Founders assume the US version is a translation job: strip the other languages, swap the CE mark for an FDA clearance reference, done.

It is not a translation job. The US IFU is a separate labelling artifact with its own content expectations under US labelling rules. The indications for use statement is not the same as the MDR intended purpose — the two need to be consistent but phrased according to each system's conventions. US warnings and precautions follow a different structure. The UDI uses the GUDID format rather than EUDAMED. The manufacturer and US Agent information must appear on the labelling in the US-required places. Promotional claims that are compliant in the EU may cross FDA promotional rules.

The fix is simple and cheap if done early: have the US IFU drafted by someone who knows US labelling, not by a translator. The cost of the draft is small. The cost of a relabelling campaign after an FDA inspection finding is not.

Mistake 4 — Treating the US Agent as paperwork

Foreign manufacturers that export devices to the US must designate a US Agent who serves as the FDA's point of contact in the United States. The US Agent receives FDA communications, assists with FDA inspections, and acts as the first line of contact for regulatory correspondence. This is a legal requirement for foreign manufacturers, not an administrative courtesy.

EU startups frequently treat the US Agent appointment as a checkbox and pick whichever name a consultant provides or whichever service sells the cheapest annual contract. That works until it does not. When the FDA sends a communication, when an inspection is scheduled, when a post-market event triggers agency questions, the US Agent is the person the FDA actually talks to. A US Agent who does not understand your device, does not respond quickly, or does not have the regulatory literacy to triage FDA communications is a liability.

The fix is to treat the US Agent designation as a regulatory relationship, not a line item. Talk to the candidate. Understand their responsiveness, their device-category experience, and how they handle FDA correspondence. The marginal cost of a good US Agent over a cheap one is trivial compared to what a bad one costs in a real incident.

Mistake 5 — Assuming MDR clinical evidence is sufficient for the FDA

MDR Article 61 and Annex XIV require clinical evidence for every device, with methodology that follows MDCG guidance and in many cases includes systematic literature review, equivalence analysis, or clinical investigation. The CER is a comprehensive document and, for higher-risk devices, a substantial one.

Founders then assume the same clinical file satisfies the FDA. Sometimes it partially does. Often it does not. For a 510(k), clinical data may not be required at all — and in that case the question of "MDR clinical sufficiency" is moot. For a De Novo or PMA, the FDA will scrutinise clinical evidence in detail, and the agency has its own views on trial design, endpoints, statistical power, and patient population. A clinical investigation designed only for MDR Annex XIV can miss FDA-critical elements and become unusable as primary evidence in a US submission without additional work.

The fix is to design the clinical evidence plan with both regulators in mind from the beginning. If a clinical investigation is in scope, the study protocol should be reviewed by someone who knows the FDA's current expectations in your device category before enrolment starts. Retrofitting a trial for the FDA after it has run is the expensive path.

Mistake 6 — Skipping a real US distribution partner

This one is not a regulatory mistake but it kills US market entry just as reliably. EU startups secure FDA clearance, register the establishment, ship product across the Atlantic, and then wait for sales to materialise in a market where they have no commercial footprint, no hospital relationships, no reimbursement strategy, and no feet on the ground. The clearance is real. The revenue is not.

The US medical device market is large, fragmented, and relationship-driven. Hospital procurement cycles are long. GPO contracts matter. Reimbursement coding matters. Field sales and clinical support matter. None of this is covered by an FDA clearance. A CE-marked device with FDA clearance and no US distribution partner is a product with the right paperwork and no path to a customer.

The fix is to treat US distribution as a parallel workstream to US regulatory. Identify distributors or a direct commercial plan before clearance. Understand reimbursement and coding early. Budget for US commercial presence as seriously as you budget for regulatory submissions. The two have to land together or neither lands at all.

Mistake 7 — Ignoring the FDA's quality system expectations

The FDA's quality system framework is codified at 21 CFR Part 820 and is moving toward closer alignment with ISO 13485 under the Quality Management System Regulation harmonisation initiative. EU manufacturers who already operate an EN ISO 13485:2016+A11:2021 QMS have done most of the underlying work, but they should not assume that ISO 13485 certification alone satisfies the FDA's expectations.

The gap is in the implementation details: design controls documentation depth, complaint handling and MDR reporting to the FDA, CAPA process specifics, and the FDA's inspection expectations, which differ from a Notified Body surveillance audit. A startup that has built its QMS carefully against ISO 13485 has done the foundation. The gap analysis against the FDA's framework needs to be done early and deliberately, not discovered during a pre-submission review or a real inspection.

The fix is to do the gap analysis while the QMS is still malleable. It is a small exercise compared to the rework of retrofitting a mature QMS to cover a Part 820 finding.

The Subtract to Ship angle

Subtract to Ship for US market entry means cutting the false starts, not the work. The pattern we see with EU founders who handle the Atlantic crossing well is not "do less FDA work" — it is "do the right FDA work early instead of the wrong FDA work late." The seven mistakes above are all false starts. Each one produces visible activity without producing progress. Each one has a cheaper, earlier alternative.

The underlying regulatory work still has to happen: the QMS alignment, the predicate selection, the US labelling, the US Agent designation, the clinical strategy, the distribution plan, the quality system gap analysis. What Subtract to Ship removes is the duplication, the speculative work, and the activities that exist because the two tracks were planned in isolation instead of together. One intended purpose, drafted once with both regulators in mind. One QMS, built to cover both frameworks from day one. One clinical evidence plan, designed for both regulators where possible. Two submissions, prepared in parallel with specialists on each side. Zero wasted effort.

Reality Check — Where do you stand on these seven mistakes?

  1. If an investor asked you today how your CE mark helps you in the US, could you answer accurately in one sentence?
  2. Have you identified candidate predicate devices, and can you defend the choice under skeptical review?
  3. Is your US IFU a separate document drafted by someone who knows US labelling, or a translation of the EU IFU?
  4. Do you know who your US Agent is, how responsive they are, and how they handle FDA correspondence?
  5. Is your clinical evidence plan designed to serve both MDR Annex XIV and FDA expectations, or only one?
  6. Do you have a concrete US distribution plan with named partners or a direct commercial strategy, beyond the clearance itself?
  7. Have you done a deliberate gap analysis between your ISO 13485 QMS and the FDA's quality system framework?

If you answered fewer than four of these clearly, the US is not a plan yet. It is a wish list.

Frequently Asked Questions

Does the CE mark count for anything in the US? Not as regulatory recognition. The FDA does not accept CE marking as evidence of compliance with US requirements. Underlying technical data produced for CE (testing, risk management, clinical data) can often be reused as supporting evidence in an FDA submission, but the submission itself is separate and must meet FDA rules.

What happens if I pick the wrong predicate for my 510(k)? Best case, the FDA pushes back during review and asks for additional data or a different predicate, which adds months to the timeline. Worse case, the submission is rejected and the work has to start again. Predicate selection is the place to invest in a US regulatory specialist before drafting, not after rejection.

Can I use the same IFU in the EU and the US? No. The two are structured against different labelling rules, use different UDI formats, carry different conformity marks, and have different conventions for indications, warnings, and promotional claims. Draft the US IFU as a separate document from the start.

Is a US Agent optional for an EU manufacturer? No. Foreign manufacturers that export devices to the US must designate a US Agent who serves as the FDA's point of contact. It is a legal requirement, not an optional service.

Will my MDR clinical investigation work for an FDA submission? Sometimes partially. The FDA has its own views on trial design, endpoints, and statistical analysis, and an investigation designed only for MDR Annex XIV can miss elements the FDA considers critical. Plan the clinical strategy with both regulators in mind before enrolment starts.

Do I need a US distribution partner before FDA clearance? You need the plan before clearance, even if the contracts close after. A clearance without a commercial path is paperwork without revenue.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices (MDR), Article 2(1), Article 2(12), Article 20 (CE marking), Article 61 and Annex XIV (clinical evaluation). Official Journal L 117, 5.5.2017.
  2. EN ISO 13485:2016 + A11:2021 — Medical devices — Quality management systems — Requirements for regulatory purposes.
  3. U.S. Food and Drug Administration — Center for Devices and Radiological Health (CDRH), public guidance and regulatory framework references. Cited at the general framing level in this post.
  4. U.S. Code of Federal Regulations, Title 21, Part 820 — Quality System Regulation (medical device QMS). Referenced at the general framework level; QMSR harmonisation initiative noted.

This post is part of the FDA & International Market Access series in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. A note on the limits of our expertise: Tibor's authority is the EU MDR and the Notified Body system. For specific FDA pathway questions on your device — predicate selection, submission drafting, US Agent relationships, Part 820 gap analysis, or US distribution strategy — work with a US regulatory specialist and a US commercial partner who practice inside that system daily. This post is the orientation. The specialists are the execution.