EU MDR remediation: why technical files keep coming back

One device, one story

Under the MDR, the technical documentation set out in Annexes II and III has to show that the device meets the general safety and performance requirements (GSPRs) in Annex I. A Notified Body reviewer reads it as one argument. When the intended purpose in the instructions for use differs from the one in the clinical evaluation, or a claim in the marketing material has no support in the clinical data, the argument breaks, and the file comes back.

That is why so many nonconformities look small on their own but keep recurring. The underlying problem is consistency across documents, not any single document.

The gaps that come up most often

Intended purpose and claims that drift

The intended purpose, indications, contraindications, target population and performance claims should read the same everywhere: labelling, instructions for use, risk management file, clinical evaluation and PMS plan. Drift between versions is common after years of updates under the old directives.

A GSPR checklist without traceability

A checklist that marks each requirement “applicable, complies” is not enough. Each line needs to point to the specific standard, test report or document section that shows conformity, and explain why a requirement does not apply where that is claimed.

Clinical evaluation that cannot carry the claims

Clinical evaluation under Article 61 and Annex XIV is where files stall most often. Typical findings:

• Equivalence claimed without showing technical, biological and clinical equivalence in the depth MDCG 2020-5 describes, or without the access to the equivalent device’s data that the MDR requires for some devices.
• No clear description of the state of the art, so the benefit-risk conclusion has nothing to be measured against.
• Clinical data that does not cover every indication, population or claim in the labelling.
• For legacy devices, gaps against what MDCG 2020-6 describes as sufficient clinical evidence.

Notified Bodies assess the clinical evaluation in a structured way. Reading MDCG 2020-13, the template for their clinical evaluation assessment report, shows what a reviewer will check.

PMS that does not close the loop

The PMS plan, PSUR or PMS report, and post-market clinical follow-up (PMCF) plan or justification should feed back into risk management and the clinical evaluation. A generic PMS plan, a PSUR that summarizes data without drawing conclusions, or a PMCF justification that does not address the residual gaps in clinical evidence will all draw findings. MDCG 2022-21 sets out what is expected of a PSUR.

Risk management in isolation

Risks, risk controls and residual risks should be traceable to the design, the labelling warnings, the clinical evaluation and the PMS data. A risk file that has not been updated with post-market information is a frequent finding.

A practical order for remediation

1. Fix the intended purpose and claims first. Agree one version and apply it everywhere. Every later document depends on it.
2. Map the gaps against the GSPRs. Build or repair the traceability so each requirement points to evidence.
3. Close the evidence gaps. Clinical data, testing or literature, scoped to the claims you actually make. Narrowing a claim can be faster than generating new data.
4. Rewrite the summaries. Clinical evaluation report, PSUR and PMS report, once the evidence underneath them is stable.
5. Answer each nonconformity on its own terms. Root cause, correction, corrective action and the evidence that closes it, with references the reviewer can find quickly.

Where we help

We support manufacturers with MDR gap assessments, technical documentation, clinical evaluation and PMS, and with preparing responses to Notified Body nonconformities. We are not a Notified Body and cannot predict or promise a certification outcome. See EU MDR remediation.