A legacy device transitioning from MDD to MDR cannot carry its old clinical evaluation report across the line unchanged. The CER has to be upgraded to meet MDR Article 61 and Annex XIV Part A, which means a new clinical evaluation plan, a re-appraisal of the existing literature against stricter criteria, a tighter equivalence argument aligned with MDCG 2020-5, an Annex XIV Part B post-market clinical follow-up plan, and a documented gap analysis that shows exactly which MDD-era evidence still holds and which has to be rebuilt. The deadlines under Regulation (EU) 2023/607 determine how much time is left to do this work.

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

TL;DR

  • Legacy devices certified under MDD 93/42/EEC or AIMDD 90/385/EEC need their clinical evaluations rebuilt to MDR Article 61 and Annex XIV Part A standards before the Notified Body will issue an MDR certificate.
  • The MDD-era CER almost always falls short in three places: the equivalence argument, the PMCF plan, and the appraisal criteria for the literature review.
  • MDCG 2020-5 (April 2020) raised the bar on equivalence claims. Equivalence arguments that passed under MEDDEV 2.7/1 Rev 4 often do not pass under MDR.
  • Regulation (EU) 2023/607 extended the transition deadlines to 31 December 2027 for Class III and Class IIb implantable legacy devices and to 31 December 2028 for Class IIb non-implantable, Class IIa, and Class I (sterile/measuring) legacy devices. The clinical evaluation upgrade must be finished well before these dates.
  • The right first step is a structured gap analysis between the legacy CER and the current MDR requirements. Everything else flows from the gaps identified.

Why MDD-era CERs do not survive MDR review

A legacy device sits in an uncomfortable place. It has a real history on the market, real users, real clinical experience, and a CER that was signed off by a Notified Body under the old Directives. None of that disappears. But none of it automatically satisfies MDR Article 61 either, and that is where founders of companies carrying legacy portfolios get caught.

The MDR did not just rename clinical evaluation. It tightened what counts as sufficient clinical evidence, what counts as equivalence, and what has to happen after CE marking. A CER written to MEDDEV 2.7/1 Rev 4 under MDD was built against a different bar. When a Notified Body assesses that same device for MDR certification, the clinical evaluation is re-read against the new bar, and the gaps that were acceptable under the Directives suddenly are not.

Tibor sees the pattern in every legacy transition project. "Manufacturers come in with a CER they are proud of. It passed every MDD surveillance audit. They assume it will carry over. It will not, not without work. The Notified Body is not being difficult. The rules changed. The CER has to change with them."

What changed between the MDD CER and the MDR CER

The structural change is that clinical evaluation under MDR is defined in Article 2(44) and Article 61 as a systematic and planned process that runs continuously across the lifecycle of the device. Annex XIV Part A specifies the process. Annex XIV Part B specifies post-market clinical follow-up that feeds back into the evaluation. The CER is the output of that process at a point in time.

Under MDD, the same general idea existed in MEDDEV 2.7/1 Rev 4 (June 2016), but the obligations were softer, the equivalence route was more forgiving, and the PMCF expectations were less prescriptive.

The practical changes a legacy CER has to absorb are:

  • A clinical evaluation plan (CEP) is required up front. Annex XIV Part A paragraph 1 requires the CEP to define scope, intended purpose, target population, clinical benefits, acceptance criteria, data sources, appraisal methodology, and analysis methodology before data is collected. Many MDD-era CERs do not have a CEP at all, or they have a thin one reverse-engineered after the literature review was done.
  • Appraisal criteria must be pre-specified, not retrofitted. MDR-era Notified Bodies want to see that the criteria for including, weighting, and rejecting literature were set before the literature was screened. A CER that shows signs of appraisal criteria being written to match the data that was found will be challenged.
  • Equivalence is much tighter. MDCG 2020-5 raised the bar on technical, biological, and clinical equivalence, and narrowed the space where equivalence to a competitor device can be claimed without contractual access to the competitor's technical documentation.
  • PMCF is now a continuous obligation, not a suggestion. Annex XIV Part B requires an active, structured PMCF plan and a PMCF evaluation report that feeds the next CER update. A legacy PMCF plan that was essentially "passive complaint monitoring" will not pass.
  • Benefit-risk and state-of-the-art have to be addressed explicitly. The MDR requires the clinical evaluation to consider currently available alternative treatment options and the state of the art. MDD-era CERs often did not have a dedicated state-of-the-art section at the depth an MDR reviewer now expects.

The gap analysis: the only sensible first step

Before anything is rewritten, the legacy CER has to be read against current MDR requirements and every gap documented. Without this gap analysis, the upgrade becomes a panicked rewrite where time and money go into the wrong sections.

The gap analysis is a structured comparison. On one side, the legacy CER, the MDD technical file, the PMS data collected since the device was first placed on the market, and any complaint and vigilance records. On the other side, MDR Article 61, Annex XIV Part A, Annex XIV Part B, MDCG 2020-5, and the current state of the clinical literature for the device and its comparators.

The output is a gap list. For each gap, the analysis records what the MDR requires, what the legacy CER currently provides, what the delta is, and what work would close it. Some gaps are editorial (the CER has the content but not in the MDR-expected structure). Some are analytical (the literature is there but the appraisal criteria are missing or were not pre-specified). Some are substantive (the equivalence argument does not hold under MDCG 2020-5 and the equivalent device cannot be used). The editorial gaps are cheap. The substantive ones are where the budget and the timeline live.

A disciplined gap analysis also identifies what the legacy CER already does well. MDD-era clinical evaluations built on years of real-world use often contain PMS signal, vigilance experience, and clinical history that is genuinely valuable under MDR. This content does not need to be thrown out. It needs to be re-framed, re-appraised against the new criteria, and integrated into the new CEP.

Regulation (EU) 2023/607 and how it shapes the timeline

Regulation (EU) 2023/607 amended Article 120 of the MDR and extended the transition periods for legacy devices. For Class III and Class IIb implantable legacy devices, the deadline is 31 December 2027. For Class IIb non-implantable, Class IIa, and Class I devices requiring Notified Body involvement (sterile or measuring function), the deadline is 31 December 2028. See our walkthrough of the MDR amendment (EU) 2023/607 for the full conditions and the exact class-by-class deadlines.

What this means for the clinical evaluation upgrade is that the real internal deadline is not the 2027 or 2028 date on the certificate. It is the date the Notified Body needs a complete MDR technical file to finish its assessment before the legacy deadline hits. Most Notified Bodies estimate 12 to 18 months from submission to certificate for a straightforward review. Queues can add more. A Class III legacy device with a 31 December 2027 transition deadline is already, in April 2026, inside the window where the clinical evaluation upgrade needs to be running, not planned.

The transition provisions also impose a separate continuous-compliance condition: the device must not undergo significant changes in design or intended purpose during the transition. That constraint affects the clinical evaluation strategy directly. Any change made to the intended purpose to help the new CER (for example, narrowing indications to match the available literature) has to be weighed against the risk of triggering the significant-change rule and losing the transition protection entirely. This is a place where expert judgement is unavoidable.

The common legacy gaps, in order of cost

Across the legacy transition projects Tibor has supported, the same gaps repeat. In rough order of how expensive they are to close:

Gap 1 — No pre-specified appraisal criteria for the literature. Cheapest to fix if the literature itself is still adequate. The appraisal criteria are written into a new CEP, the existing literature is re-screened against them, and the weighting is documented. Editorial work, but non-trivial.

Gap 2 — Missing or thin state-of-the-art section. The legacy CER did not go deep on alternative treatments or the current clinical landscape. This is a writing and literature task, not a new investigation task.

Gap 3 — Weak PMCF plan under Annex XIV Part B. The MDD-era PMCF was passive. The MDR-era PMCF has to be active, structured, and feasible. A new PMCF plan is written, aligned with the risk class and the nature of the device, and the existing post-market data is re-framed as PMCF input where possible.

Gap 4 — Equivalence argument that does not hold under MDCG 2020-5. The legacy CER relied on equivalence to a device the manufacturer does not have contractual access to, or the equivalence claim did not fully address technical, biological, and clinical dimensions. This gap is expensive because it often forces either a new equivalence partner, a rebuild of the evidence from literature and PMS data alone, or a new clinical investigation.

Gap 5 — Insufficient clinical data at the current MDR bar. The rarest and most expensive gap. The legacy evidence, even re-appraised, does not reach the level of clinical evidence required under MDR Article 61 for the device class and intended purpose. The only options are narrowing the intended purpose, building additional evidence through PMCF bridging studies, or running a new clinical investigation under EN ISO 14155:2020+A11:2024.

Re-appraising old literature: what survives, what does not

The literature in a legacy CER is usually the single largest body of content, and it is also usually the content most affected by the re-appraisal exercise. The old appraisal was done under MEDDEV 2.7/1 Rev 4 criteria, often without a documented pre-specified appraisal framework. The MDR re-appraisal asks a stricter question: does this paper, on its own merits, support the specific clinical claim it is being used to support, given a pre-specified appraisal method and weighting?

Three things typically happen in the re-appraisal. Some papers survive unchanged because they were high-quality, directly relevant, and properly weighted the first time. Some papers fall out because, under honest re-appraisal criteria, they do not actually support the claim they were cited for. And some papers get re-purposed, moving from supporting one claim to supporting another that they actually fit better.

The honest re-appraisal is uncomfortable because it sometimes exposes that a legacy CER was leaning on literature that did not really support what it was cited for. This is the moment where founders are tempted to paper over the gap. Tibor's rule on this is absolute. If the literature does not support the claim, the claim has to change, the evidence has to be supplemented, or the intended purpose has to be narrowed. Patching the appraisal to hide the gap fails at the next Notified Body review and damages credibility permanently.

PMCF integration: turning post-market history into MDR evidence

A legacy device has one genuine advantage a new device does not: years of real-world post-market data. Complaint records, service reports, vigilance cases, customer surveys, sales and usage volume, clinical experience from users. Under MDD, much of this sat in PMS files and was reviewed for vigilance triggers. Under MDR, with Annex XIV Part B requiring active PMCF that feeds back into the CER, this post-market history becomes clinical evidence if it is re-framed correctly.

The integration work has two sides. First, retrospective: the existing post-market data is screened for clinically relevant signal, re-structured as PMCF input, and fed into the upgraded CER. Second, prospective: a new PMCF plan is written that defines what data will be collected going forward, at what cadence, by what methods, and how it will feed into the next CER update. The retrospective work reuses what exists. The prospective work makes sure the next CER cycle is built on an active foundation.

For legacy devices with strong post-market history, this integration can carry substantial clinical evidence weight and reduce the need for new clinical investigations. For legacy devices whose post-market data collection was weak or passive, the retrospective side is thinner and the prospective side has to be designed more aggressively to close the gap in time.

A typical timeline

For a straightforward legacy transition project where the gap analysis identifies mostly editorial and analytical gaps, the clinical evaluation upgrade runs roughly six to nine months: one to two months for the gap analysis and new CEP, two to three months for the literature re-appraisal and state-of-the-art update, one to two months for the PMCF plan and retrospective integration, one to two months for the CER rewrite, and a buffer for review and iteration. The Notified Body review then adds 12 to 18 months on top.

For a project where the gap analysis exposes a broken equivalence argument or an insufficient clinical evidence gap, the timeline extends by whatever the evidence-rebuild takes, which can be one to two years on top. This is why the gap analysis has to happen first and happen honestly. A company that discovers a substantive gap 12 months before the transition deadline is in a very different position than one that discovers it 30 months before.

The Subtract to Ship angle

The Subtract to Ship framework applied to legacy transition is a disciplined refusal to rewrite sections of the CER that do not actually have a gap. The gap analysis is the subtraction tool. Every section of the legacy CER that already meets MDR requirements stays as-is, with the CEP referencing it. Every section with a real gap is rewritten to close the gap and no further. The temptation to "rewrite everything to be safe" is expensive and usually counter-productive — it introduces inconsistency, wastes the lifecycle evidence already in the legacy CER, and burns runway on sections that were fine.

Subtraction also applies to the literature. Do not pad the re-appraised literature set with marginal papers to inflate the evidence volume. A tightly curated, honestly appraised literature set that cleanly supports the specific claims is stronger than a bloated set that obscures the argument. This is the same discipline as the Evidence Pass in new-device clinical evaluation: cheapest legitimate pathway first, no duplication, every piece of evidence earning its place.

Reality Check — Where do you stand?

  1. For each legacy device in your portfolio, have you identified the specific MDR transition deadline under Regulation (EU) 2023/607 based on its risk class?
  2. Have you completed a structured gap analysis between the legacy CER and the current requirements of MDR Article 61 and Annex XIV Part A?
  3. Does your legacy CER have a pre-specified clinical evaluation plan, or was the CEP written after the fact to match the data that happened to be in the file?
  4. If the legacy CER relies on equivalence, does that equivalence argument still hold under MDCG 2020-5, including sufficient levels of access to the equivalent device's technical documentation?
  5. Has the existing literature been re-appraised against pre-specified criteria, and do you know honestly which papers survive and which do not?
  6. Does your PMCF plan meet Annex XIV Part B, or is it a passive complaint-monitoring arrangement inherited from the MDD era?
  7. Have you mapped the retrospective post-market data of the legacy device into the new CER as PMCF input?
  8. Working backward from the legacy transition deadline and the Notified Body review time, do you have enough calendar left to finish the clinical evaluation upgrade without compressing the honest work?

Frequently Asked Questions

Can a legacy device keep its MDD clinical evaluation report under MDR? No. The Notified Body will assess the clinical evaluation against MDR Article 61 and Annex XIV Part A, not against the MDD. Parts of the legacy CER content can be reused, but the CER as a whole has to be upgraded — new CEP, re-appraised literature, tightened equivalence, active PMCF plan, explicit state-of-the-art section.

What is the biggest difference between an MDD CER and an MDR CER? The biggest structural difference is that clinical evaluation under MDR is a continuous, planned process with a pre-specified CEP and a PMCF feedback loop defined in Annex XIV Part B. The biggest practical difference for legacy devices is that equivalence claims are much tighter under MDCG 2020-5 and that PMCF is an active obligation, not a passive one.

Does Regulation (EU) 2023/607 give me more time to upgrade the CER? It extends the transition deadline for the device certification, not the substantive requirements. The clinical evaluation still has to meet MDR Article 61. The extension gives you more calendar to do the upgrade properly, but the extension is conditional on the transition requirements being met and the device being actively worked through MDR conformity assessment.

If the legacy literature does not support the claims under honest re-appraisal, what are my options? Three options: narrow the intended purpose so the surviving literature fully supports the narrower claims, build additional evidence through PMCF activities that target the uncovered claims, or run a new clinical investigation under EN ISO 14155:2020+A11:2024 for the specific gap. Patching the appraisal is not an option.

How much of the old PMS data can be used as PMCF evidence under MDR? Most of it, if it is re-framed correctly. Complaint data, vigilance data, real-world use data, and customer feedback that was collected under the MDD-era PMS can be integrated into the MDR CER as PMCF input, provided the appraisal is honest and the data is relevant to the specific clinical claims being supported. A new prospective PMCF plan still has to run on top of the retrospective integration.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, Article 2(12) (intended purpose), Article 2(44) (clinical evaluation), Article 61 (clinical evaluation), Article 120 (transitional provisions), Annex XIV Part A (clinical evaluation), Annex XIV Part B (post-market clinical follow-up). Official Journal L 117, 5.5.2017.
  2. Regulation (EU) 2023/607 of the European Parliament and of the Council of 15 March 2023 amending Regulations (EU) 2017/745 and (EU) 2017/746 as regards the transitional provisions for certain medical devices and in vitro diagnostic medical devices.
  3. MDCG 2020-5 — Clinical Evaluation — Equivalence: A guide for manufacturers and notified bodies, April 2020.
  4. MEDDEV 2.7/1 revision 4 — Clinical Evaluation: A Guide for Manufacturers and Notified Bodies under Directives 93/42/EEC and 90/385/EEC, June 2016 (legacy guidance, referenced for structural comparison; MDR text takes precedence where they diverge).
  5. EN ISO 14155:2020 + A11:2024 — Clinical investigation of medical devices for human subjects — Good clinical practice.
  6. EN ISO 14971:2019 + A11:2021 — Medical devices — Application of risk management to medical devices.

This post is part of the Clinical Evaluation & Clinical Investigations series in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. Legacy transitions are where regulatory history meets the current rulebook, and the clinical evaluation upgrade is usually the part where a smart gap analysis saves the project — or its absence sinks it.