Within the medical device industry, the European Medical Device Regulations (MDR) 2017/745 has introduced new challenges, increasing the scrutiny of the regulatory documentation and imposing stricter requirements for approving medical devices in the EU market. The MDR emphasizes the need for substantial clinical evidence to demonstrate conformity with the relevant general safety and performance requirements and a favourable benefit-risk profile. One way to gather such clinical evidence is through a high quality systematic review. In this blog post, we highlight the key roles of literature reviews, offer practical advice to improve the methodology and explore some of the most common deficiencies identified by Notified Bodies when assessing systematic literature reviews.
Key roles of the literature review
High-quality literature reviews now play a central role under the MDR, as they help identify clinical data and bridge gaps in clinical evidence. For certain devices, most of the clinical evidence comes from clinical data identified in literature reviews, making it essential to conduct these reviews meticulously. Literature reviews are complex and time-consuming, requiring a blend of technical, analytical, and writing skills. Therefore, they are best performed with an experienced medical writer to ensure efficiency.
Let us go over the various processes the literature review contributes to. They are summarized in the infographic below.
Establishing the state of the art
A systematic literature review provides a comprehensive overview of current knowledge and the state of the art in the medical field, outlining the natural course of the medical condition. It should also present the device under evaluation, comparing it to benchmark devices and other treatment options and include current clinical guidance for treating the medical condition.
Demonstrating safety and performance
The main goal of the literature review is to gather safety and performance data from scientific literature and assess how the device under evaluation compares to similar or alternative devices.
Reviewing the literature can also provide valuable insights into safety and performance parameters assessed for similar devices, which can be further implemented into the manufacturer’s own clinical investigations.
Identifying risks and post-market surveillance
A literature review can uncover new risks and adverse events associated with similar devices, prompting updates to the risk management documentation. In such cases, the manufacturer must develop appropriate mitigation strategies to ensure that the benefit-risk of the device remains acceptable.
As part of the post-market surveillance, manufacturers should regularly update the literature review for continuous monitoring of the safety, identify potential new risks, and stay compliant with the post-market surveillance obligations laid out in the MDR.
A Critical Component of the Clinical Evaluation
The clinical evaluation report requires robust data to demonstrate that the device performs as intended and that its benefits outweigh any risks.
The systematic literature is a critical component of this clinical evaluation. Although it is recommended to have a separate literature review plan and report, these are ultimately summarized in the clinical evaluation plan and report, making the literature review a significant part of the clinical evidence.
The clinical evaluation, along with the systematic literature review, should be conducted early in the manufacturing process, as it can provide valuable insights into design features.
Since the systematic literature review represent a substantial portion of the clinical data, it is closely examined by the Notified Bodies (NB) and is often the object of non-conformities.
Methodology for conducting literature reviews under the MDR
The literature review methodology should be transparent and thoroughly documented. A poorly documented or non-reproducible review can raise concerns from the NB and delay the path to CE-marking. The Meddev 2.7/1 revision 4 outlines the principles of clinical evaluation and provide details on the literature review process. Introduced in 2016 under the Medical Device Directive 2007/47/EC (MDD), it aligns with the MDR requirements and remains the main reference for conducting literature reviews as part of the clinical evaluation.
The literature review process consists of the following steps:
1. Define the keywords, search terms and overall plan
The first step is defining a set of keywords and building search terms. Using a structured process, such as the PICO framework, ensures a methodological and systematic approach.
Formulate a list of research questions related to the state of the art, safety, and performance of the device. These questions will guide the process, helping identify keywords, and for building an exhaustive state of the art.
Several goal-oriented searches should be planned, such as searches for the state of the art, clinical guidance, and for the safety and performance of the device under evaluation or similar and alternative devices. For each planned search, define a search period and carefully select at least two databases. Common sources include PubMed, Cochrane Reviews, Cochrane CENTRAL, Clinicaltrials.gov, Google Scholar, Medscape, UpToDate, website of professional societies.
Plan the analysis of the safety and performance data, establishing which parameters will be collected and whether the analysis will be merely qualitative or also quantitative.
Document the methodology in a literature review plan, including keywords, search terms, research questions, databases selected, selection process, appraisal, and analysis plan. You can use Lorentis’ Literature Review Plan, available on our website to ensure a comprehensive plan.
2. Execute the searches
Conduct the searches as outlined in the plan.
3. Identify relevant publications
Document the selection process, adhering to the selection criteria set out in your literature review plan. Provide clear reasons for excluding any publications, and eventually create a PRISMA flowchart to show the number of records identified, included, excluded and the reasons for exclusions.
4. Appraise the clinical data
Appraise all selected publications presenting safety and performance data to determine their scientific validity and relevance to the clinical evaluation. Several methods exist for appraising clinical data, and a common method is outlined in the IMDRF MDCE WG/N56FINAL:2019 document, which defines criteria for assessing suitability and data contribution. Using this system, assign a weight to each publication, defining its importance in the demonstration of safety and performance.
5. Compile the literature review report
The literature review report should detail the methodology used, any deviations from the plan, the state of the art in the medical field and an analysis of the safety and performance data encountered in the scientific literature.
Using the narrative reviews, meta-analyses and clinical guidelines retrieved by the searches, the state of the art should describe the medical condition(s) with the disease course, severity and affected population. It should also present the different treatment options available for the disease, defining the intended use of the device under study, the experience with similar devices and alternative treatment options with the advantages and drawbacks of their use, and outline the differences between them. Summarize current clinical guidance and the standard of care for the medical condition and assess whether the device under study or similar devices can be considered standard of care or are recommended for the treatment of the disease.
Extract all the safety and performance data encountered in the selected publications, along with the general characteristics of the studies described (e.g., type of study, number of patients, age, indication, inclusion and exclusion criteria). Summarize these data in tables describing the relevant safety and performance parameters and their range of values found in the literature.
Finally, draw a conclusion on the device under evaluation: How does it compare to similar and alternative devices in terms of safety and performance? Can the literature review support the claims made for this device? What are the benefits and risks of using this device?
You can use the Lorentis Literature Review Report Template available on our website to guide you.
Common deficiencies identified by Notified Bodies
The Medical Devices Coordination Group (MDCG) published a template for the Clinical Evaluation Assessment Report (CEAR), which is prepared by the NB to document its assessment of the clinical evaluation and related documents. This MDCG 2020-13 guidance provide valuable information for manufacturers by outlining the key aeras of focus for NBs when assessing systematic literature reviews.
Below are common non-compliances of systematic literature reviews observed by NBs.
Missing or poorly defined search strategy
One of the most frequent deficiencies involves an incomplete or missing search strategy. NBs expect a transparent and detailed search strategy that can be reproduced.
Common issues include:
- Literature search strategy missing
- Literature search not reproducible
- Only one database selected, introducing a risk of bias.
- Literature search period not defined, making it unclear whether the literature review reflects current knowledge.
- Literature search did not follow an objective search method (such as PICO), resulting in an incomplete review of available literature.
Inadequate inclusion and exclusion criteria
Another major issue commonly encountered by NBs are when the inclusion and exclusion criteria are unclear, incomplete, or not applied consistently. As a result, relevant study may be arbitrarily excluded.
Common issues include:
- Inclusion and exclusion criteria missing or poorly defined
- Justification for excluded publication missing
Appraisal of clinical data lacking or inadequate
NBs frequently come across deficiencies relating to the appraisal of the clinical data with either a missing or not comprehensive appraisal.
Common issues include:
- Appraisal plan missing
- No systematic method for assessing the quality and relevance of the included studies.
- Failure to assign weight to studies
Failure to establish a comprehensive state of the art
Many manufacturers fail to provide a complete state of the art describing the current knowledge for the medical field and medical device in question, including alternative treatment options and benchmark devices.
Common issues include:
- The state of the art does not cover the full range of alternative therapies or comparable devices.
- No clinical guidelines or standards or care are referenced.
- Incomplete analysis of how the device under evaluation compared with competing devices in terms of safety and performance.
Literature review plan or report not provided
As outlined in the MDCG 2020-13, NBs expect a complete literature review documentation including the literature review plan, the literature review report, a list of all retrieved publications, a list of all excluded publications with reasons for exclusion, and the full text of the selected publications.
Common issues include:
- Missing or incomplete literature review plan and report.
- Failure to outline the methodology and scope of the review in the plan.
- Deviations from the original plan not documented or justified.
As highlighted in this blog post, careful planning, documentation, and a rigorous methodology ensures a transparent literature review process that meets the NB’s requirements and contributes to the smooth approval of your medical device.
Lorentis specializes in delivering high-quality medical writing services. From setting up the searches and methodology, to the conduction of the literature review and writing of the plan and report, our experienced medical writers will ensure an MDR-compliant literature review, helping reduce time-to-market.
Lorentis works with Lindexer software to manage its systematic literature reviews.
