Literature Search Protocol for EU MDR: Structure, Template, and When to Update
A literature search protocol is the written plan, fixed before any searching begins, that defines exactly how a literature search will be run. It sets out the research questions, databases, search strings, eligibility criteria, screening and appraisal methods, and how results will be documented. Under EU MDR it matters because notified bodies judge a literature review on whether it can be reproduced, and the protocol is what makes it reproducible. It is also what you reuse at every CER update, so a good one saves work for the life of the device. For the full review process the protocol feeds into, see our step-by-step EU MDR literature review guide.
Getting the long-awaited CE mark is certainly something to celebrate, so if you got it, congratulations! Now you just have to make sure to keep it!
While the preparation for the device launch is difficult, the real work comes after the device is placed on the market, when the CER has to be kept up to date for as long as the device is sold. There are tons of stuff to be done; analyses to run, clinical data to collect, and reports to update periodically.
One of the big things to take care of is the literature search or literature review. The literature search takes up much of the workload of the CER process. So, it would be great if we could do it once and be done with it. However, the field of medicine is always changing. Whether there are new devices being invented every day or new information regarding existing technology, it all translates to one thing: an update in the literature review.
Every update cycle requires the same rigorous standards as the initial submission, which our EU MDR literature review guide walks through step by step.
In this article, we will go through the common questions people have about the process and some tips for you.
What Goes in a Literature Search Protocol: Sample Structure
A complete protocol covers scope, research questions, sources, search strings, eligibility criteria, screening, appraisal, extraction, reporting, and an update plan, all fixed and version-controlled before searching.
There is no single mandatory template, but notified bodies expect the elements below. Many manufacturers attach the protocol to the clinical evaluation plan required by Annex XIV Part A, so that the plan and the search stay in step.
| Section | What it should contain |
| 1. Document control | Version, date, authors and reviewers, their qualifications, and declarations of interest (as MEDDEV 2.7/1 Rev 4 expects of evaluators) |
| 2. Scope and objectives | Device, intended purpose, indications, target population, and the clinical claims the search must support; cross-reference to the clinical evaluation plan |
| 3. Research questions | Separate questions for the state of the art and for the device and any equivalent device, framed with PICO (population, intervention, comparator, outcome) |
| 4. Sources | Databases (e.g. PubMed, Embase, Cochrane Library) and the reason for each; supplementary sources such as ClinicalTrials.gov, registries, vigilance databases, and grey literature |
| 5. Search strategy | The full search string for each database, including Boolean logic and subject headings (MeSH, Emtree), date range, language, and any filters, each with a justification |
| 6. Eligibility criteria | Inclusion and exclusion criteria for population, study design, device relevance, outcomes, publication type, date, and language |
| 7. Screening process | Deduplication method, two-stage screening (title/abstract, then full text), number of reviewers, and how disagreements are resolved |
| 8. Appraisal method | How each study is scored for suitability and contribution, and how pivotal and supportive data are distinguished and weighted |
| 9. Data extraction | The extraction form: study design, population, device details, comparator, safety and performance outcomes, adverse events, follow-up |
| 10. Analysis and reporting | How data are synthesised, the PRISMA 2020 flow diagram, and the structure of the literature report |
| 11. Update plan | Update frequency, triggers for an unscheduled update, and how the search will be re-run for comparability |
| 12. Deviations | A log of any changes from the approved protocol, with the reason and the date |
For how each section is used in practice, see the steps in our EU MDR literature review guide.
Is Your Data Latest and Relevant?
A literature review should be updated whenever new research, practice changes, or new rules make the current version out of date, and on a regular schedule in between.
If you are wondering whether you should update the literature review of your CER, the answer lies in this question. A literature review should be updated when new research or material becomes available.
This can include everything from new findings, new theories or even hypotheses, shifts in the field practice, groundbreaking clinical trials, or the implementation of new laws or norms. It makes sense if you think about it. A literature review represents the most recent results and advancements in the subject and retains its relevance and credibility. So, as soon as the current information changes, the literature review should also change to represent it.
Updating a literature review also helps manufacturers ensure that their device design and clinical conclusions are based on the most recent information.
However, it’s not possible to redo the literature review process every day, so we do the next best thing. We update it periodically.
How Often to Update the Literature Search
High-risk or novel devices are typically updated at least annually, well-established lower-risk devices every two to five years, and any significant new safety or state-of-the-art information triggers an unscheduled update.
| Situation | Typical update frequency | Source |
| Device carries significant risks or is not yet well established | At least annually | MEDDEV 2.7/1 Rev 4, Section 6.2.3 |
| Well-established device without significant risks | Every 2 to 5 years, with justification | MEDDEV 2.7/1 Rev 4, Section 6.2.3 |
| Class III and implantable devices | PMCF evaluation report and SSCP at least annually, which in practice drives an annual literature update | MDR Article 61(11) |
| Class IIa, IIb, and III devices | Aligned with the PSUR cycle so each PSUR reflects current literature | MDR Article 86 |
An unscheduled update is warranted whenever one of these triggers occurs:
- a new serious incident, FSCA, or safety signal for your device or a similar device;
- a new harmonised standard, MDCG guidance, or clinical practice guideline that shifts the state of the art;
- a design change, new indication, or new target population;
- PMCF or PMS findings that raise a new clinical question; or
a notified body finding on the previous review.
Preparation of Literature Search and Review
Before re-running the search, index what the current review already contains, gather recent news and adverse events, and collect any reviewer comments on the last version.
The golden key to the regulatory process hides in preparation, plans, and protocols. A well-planned literature search and review strategy works like magic to pave the way for a thorough and focused search, increase the speed of the review process, and, all-around, easier evidence analysis.
The preparation process is pretty simple. The first screening is on what you already have. Create an index or use the previous one if you have one on the contents of your current literature review. Mark and tag what you can. Try to divide the contents into a few groups based on a common theme.
After you have done that, take a look at the recent news and updates of the device in question. Gather as much as you can, keeping the focus on relative and recent info. You don’t have to dig too deep; just the headlines or keywords are enough. These will serve as starting points in your literature search protocol.
Next, list all the adverse effects, complications, or clinical notes you have received so far as part of your clinical evaluation report activity. Also, enquire whether there were any comments on the last literature review from the regulatory people.
After you have enough data on the current status, start your literature review.
EU CER Literature Review Process
The review itself runs as identify, appraise, and analyse; the full step-by-step process is covered in the main guide.
The literature review process follows the same stages at every update: identifying relevant literature, appraising its quality and relevance, and analysing what it means for the device. Our step-by-step EU MDR literature review guide covers each stage in detail. What changes at an update is the starting point: you are re-running an approved protocol against a known baseline, not designing a search from scratch.
The outcome of the literature review depends on the literature search process. There are several ways to go about it, really. You can choose to dive blindly into the sea of journals and take your chances. But, for us, a systematic approach has always worked best.
Systematic Literature Search Protocol
A systematic protocol tailors the search strategy to each database, refines terms, and fixes the screening, extraction, and synthesis methods in advance.
A systematic literature review protocol means having a methodical approach that combs through the databases to find the best and most relevant literature. The outcome you get is worth the time and effort it takes to build a well-thought-out protocol.
The best protocols for a systematic review update will reflect its purpose. Choose and define inclusion and exclusion criteria based on the data you collected in the preparation step.
You will get better results for different databases by customizing the literature search strategy. Also, remember the search results depend largely on search terms, especially on platforms like Google Scholar. Term refinement in literature search in itself is a great tool to filter out the most usable literature. Focus on relevant clinical evidence and equivalent devices in the search terms.
After you go through the major scientific databases, you will have to filter them out. Have a screening and quality assessment process. Determine how you will extract the clinical data for clinical evidence at this stage. This will come in handy later while updating the clinical evaluation report.
Apart from these, you will also need to choose the ideal method of data synthesis suitable for your literature review.
For an update, keep the search strings and databases the same as the approved protocol wherever possible, and search from the date of the last search. Changing the strategy between cycles makes it hard to show the notified body what is genuinely new, so any change should be logged as a protocol deviation with its reason.
Review It All: The Favorable…and the Not So Favorable Data
The MDR requires both favourable and unfavourable data to be considered, and leaving out adverse findings catches up with you at the PMS stage.
After you are done with the first few sessions of literature searches, you should have enough to start. Start with the statistical analysis of the clinical data extracted from various sources. The literature search, when done correctly, will provide an ample amount of clinical data to go on.
The clinical evaluation process is about knowing where your strength and weaknesses lie. Sure, it is tempting to present only good data and avoid all potential adverse reactions. But you will feel the full effect at the post-market surveillance stage.
Expert Advice on Literature Search
Read the relevant MDCG guidance, check CER writer qualifications, use automation where it helps, and don’t skip grey literature.
In the field of literature review and technical documentation, pieces of advice are always welcome. Here are some of the hand-picked ones from our staff for you.
MDCG Documents
Before you complete your new and shiny literature review, go through the MDCG guidance that applies. The most relevant are MDCG 2020-13 (the clinical evaluation assessment report template notified bodies use, which covers the literature search), MDCG 2020-5 (equivalence), MDCG 2020-6 (sufficient clinical evidence for legacy devices), and MDCG 2020-7 and 2020-8 (PMCF plan and PMCF evaluation report templates). These are guidance documents from official authorities and are tremendously useful.
Hiring a clinical evaluation report writer?
If you are considering hiring someone to write your literature review, look for the specific qualifications that mark a good CER writer.
Automate the process
There are several tools at your disposal now that can help streamline the literature search. While most will help in only certain parts of the process, they are worth a try.
Grey literature
We often skip these while searching for the perfect scientific literature. However, you may find some gems within government papers or clinical trials you would otherwise miss.
FAQ
What is a literature review search protocol?
It is the written, approved plan for a literature search: the research questions, databases, exact search strings, eligibility criteria, screening and appraisal methods, and reporting approach, all fixed before searching starts so the search can be reproduced.
Is a literature search protocol required under EU MDR?
The MDR doesn’t use the word “protocol”, but Annex XIV Part A requires the clinical evaluation to follow a defined, methodologically sound procedure set out in a clinical evaluation plan, and MDCG 2020-13 directs notified bodies to check the search methodology. In practice, a documented protocol is how every manufacturer meets that expectation.
What makes a search protocol systematic?
A systematic search protocol is comprehensive (several databases plus supplementary sources), explicit (full search strings and criteria written down), and reproducible (another reviewer following it would get the same records). It also defines the screening and appraisal methods, not just the search.
Can I reuse the same literature search protocol for CER updates?
Yes, and you should. Re-running the approved protocol from the date of the last search keeps results comparable. Any changes, such as adding a database, should be logged as deviations with a reason.
Where does the literature search protocol sit in the technical documentation?
Usually as an appendix to, or a section of, the clinical evaluation plan, with the executed search and results reported in the literature review report that supports the CER.
How often should the literature search be updated?
At least annually for high-risk or not-yet-established devices, and every two to five years for well-established lower-risk devices, following MEDDEV 2.7/1 Rev 4. Class III and implantable devices also need their PMCF evaluation report and SSCP updated at least annually under Article 61(11).