MEDDEV 2.7.1 Revision 4: Guidelines for Literature Search

By Published On: January 11th, 2022Categories: EU MDR, Literature Search and Literature Review, Meddev

The MEDDEV 2.7.1 is the most important technical guidance document for literature search under the MDR in the medical device regulatory industry. Since the MDCG documents do not currently offer any concrete guidance on how the literature search should be conducted, the MEDDEV 2.7.1 revision, 4 is the primary document that provides “general guidance on a literature search and literature review protocol, key elements. The literature search and review process helps find information” on the medical device under evaluation, the equivalent devices and state of the art, including alternative examination and treatment methods to demonstrate equivalence and compliance.

 

MEDDEV 2.7.1 Revision 4

 

The most important aspects to remember when documenting the literature search are documented in Annex 5 of 2.7 1 rev 4. The MEDDEV 2.7 1 rev 4 focusses on the following sections of Literature search and review:

  • Search categories (e.g. medical device search or state of the art including clinical condition),
  • Scope of the search strategy
  • Search and review methods
  • Literature search documentation.

An important amendment in MEDDEV 2.7.1:

The literature search must be Europe-centric, emphasizing the search for European data, which is not always available in PubMed databases. Prior to this amendment, PubMed was often selected as the single source for the main search due to the probability of finding top-ranking studies in its database. Therefore, it was regarded as sufficient and was accepted.

 

The rationale for this has been described as follows: “Studies yielding negative results or user experience (…) may not qualify for publication in high impact medical journals. “One should search journals and other sources available to European users. For a Europe-centric literature search, databases such as Embase are recommended. Embase is one of the European databases recommended for a Europe-centric literature search. While PubMed allows free search and registration, access to the full texts is not always free of charge on PubMed.

 

amendment in MEDDEV 2.7 1

 

Essential Objectives of the literature search:

In accordance with the MEDDEV 2.7 1 Revision 4, the literature search should cover at least the following two objectives:

 

  1. Demonstrate state of the art for the medical device under clinical evaluation so that it can be evaluated in comparison to the state of the art
  2. Demonstrate the clinical safety and performance, clinical benefit of the medical device under evaluation to prove equivalence – Here, statements on the medical device under clinical evaluation (or equivalent devices) are required.

Documenting the objectives of the literature search:

In accordance with the guideline, the objectives of the literature search(es) must be documented. These objectives could range from:

 

  • Providing clinical data on the medical device under evaluation
  • Identifying important data for the risk management process with a focus on patient population and existing interventions
  • Providing information for the clinical evaluation of the benefit/risk profile
  • Giving an overview of the current safety specifications
  • Enabling comparison of possible side effects
  • Providing information on benchmark medical devices.

 

Depending on the objective different search terms are used.

A methodical approach is essential for the literature search process.

The sequence of literature search and selection

 

sequence of literature search and selection

 

Beginning with a suitable research question, it is crucial to select appropriate databases and formulate promising search times. In addition to databases, bibliographies targeted manual searches or catalogues from the manufacturers may also contain informative publications. It is important that search methods, search terms, strategies and particularized results be documented in detail in the final report.

How should I identify relevant publications?

Through a preliminary search in databases, it is possible to identify relevant publications by means of title, abstract, and identification of potentially relevant sources. After eliminating the publications which are not relevant, through “full texts” one can identify possibly relevant publications. Criteria for relevance should be background information, significant information regarding benefit, clinical safety and performance.

 

MEDDEV 2.7 1 revision 4 recommends the PICO method for clinical evaluation literature search – Patient/Population; Intervention; Comparison; Outcome. It is used in clinical evidence-based medicine in particular and helps with literature search, for example, when trying to find search criteria for the state of the art.

Quick Tips for an Effective Literature Search:

 

Quick Tips for an Effective Literature Search

 

Use (Boolean) operators

Boolean operators enable you to narrow down your literature search, saving you the hassle of reading non-specific literature sources. You can also use Boolean operators to expand the scope of the scientific search, especially if you don’t find enough literature sources. The Boolean operators AND, OR, and NOT can also be used to combine different search terms according to the context.

 

  • AND” – Using this operator combines terms with “AND” and filters the results for entries that contain all the search terms.
  • OR” – Using this operator, combines terms with “OR” and filters the results to show entries that contain one of the search terms.
  •  “NOT” – Eliminates entries with this search term from your search.

You can also use quotation marks and round brackets to improve the quality of the search results and get relevant and specific search results.

 

  • When you insert your search term in quotation marks, “”, the search engine will search for the search terms in that exact context and order.
  • You can also use Round brackets () to refer to a Boolean operator to terms or units.

 

Since different databases use different operators, one should check the database’s website to understand more about it. The presence or absence of brackets also affects the interpretation of the search phrase.

Exclude unrelated publications at an early stage to save time

Since a literature search often returns several hundred publications, it is important to funnel down the long list and exclude the non – relevant publications at the outset to save precious time. Based on the abstract, if publications do not relate to the state of the art or the medical device under clinical, they should be excluded from the literature review process.

Standard Mistakes during Literature Search

1. The lack of a literature search and review methodology is one of the most common mistakes when examining the systematic scientific literature review and technical documentation. A literature search protocol that does not clearly define the selected keywords, the inclusion/exclusion criteria, and the suitability criteria are challenging to recreate or reproduce. The evaluation of such a literature search is also extremely difficult.

 

2. Using literature from medical devices that do not demonstrate equivalence to the device under clinical evaluation is another common mistake. Both the MEDDEV guidance documents and the MDR explicitly state that only the literature from equivalent medical devices may be used to obtain clinical data on the device under evaluation. While the use of literature from other medical devices is allowed for the clinical evaluation reports (CER), it cannot be provided as evidence for clinical performance.

 

3. With regards to medical devices that already have the CE-mark, companies sometimes forget to look at the Post-market surveillance data. Companies are required to proactively collect Post-market surveillance data from their device as well as competitor devices. By searching for additional literature or looking into incident databases such as MAUDE, manufacturers can ensure their Post-market surveillance data is constantly updated.

 

4. Since the MDCG document 2020-13 explicitly requires the notified body to review the literature search documentation, if companies do not document the literature search completely, it will result in a non-conformity. This could potentially lead to unnecessary queries or even deviations in the audit. Companies should incorporate the following metadata into the literature search protocol – search terms, databases used, inclusion and exclusion criteria, exclusion of duplicates, Literature review procedure and documentation, search methods must be evaluated.

 

Summary

 

Summary MEDDEV 2.7.1

 

MEDDEV 2.7.1 is a useful technical guide for literature search that will help companies achieve compliance with literature search and clinical evaluation.

 

The tips in this article will help you comply with the MDR’s requirements for literature searches. One should control the range and specificity of the publications found during the literature search by using (Boolean) operators. It is also advisable to exclude irrelevant publications at an early stage. Also, in addition to the state of the art, one must at least evaluate the information on an equivalent device – thereby carrying out at least two searches.

Frequently Asked Questions

Who is MEDDEV 2.7 1 revision 4 for?

MEDDEV 2.7 1 revision 4 is for anyone doing clinical evaluations for medical devices.

Are there any upcoming updates for MEDDEV 2.7 1 revision 4?

While there are no scheduled updates planned yet, it is rumored that there is ongoing work to update the guidance to better reflect the updates to the European Medical Device Regulation, MDR.

Does MEDDEV 2.7 1 revision 4 also described the Clinical Evaluation Report?

Yes, CERs are included in the guidance. It even provides a proposed table of contents and content examples.

Want more EU MDR and Regulatory Insights?

We send weekly emails with the latest regulatory developments, templates, and strategies straight to QA/RA Professionals like you. Sign up below to get access today.