MDR Literature Search
This article will focus on the Systematic Scientific Literature Search and Review as part of the Clinical Evaluation process.
When and why is a literature search needed?
Although literature searches are commonly done as part of a device’s clinical evaluation, manufacturers should use them to obtain information throughout its entire life cycle. In agreement, the MDR mentions literature searches as a source of information in the pre-clinical evaluation, clinical evaluation, and post-market surveillance phases:
| Stage of medical device life cycle | Literature search | Purpose of the literature search |
|---|---|---|
| Pre-clinical Evaluation | Pre-clinical Literature Search Annex VII, Section 4.5.4 | Obtain information for the pre-clinical evaluation of the device |
| Clinical investigation | Background literature review Annex XV, Section 3.2 | Establish the State-Of-The-Art (SOTA), available treatment options and proposed benefits of a new device, establish the sample size of the investigation |
| Clinical Evaluation | Systematic Scientific Literature Search Annex VII, Section 4.5.5 Annex XIV, part A, Section 1 (b) | Identify ‘clinical data’ concerning the safety or performance of a device for which equivalence to the device in question can be demonstrated |
| Surveillance activities and post-certification monitoring | Screening of scientific literature Annex III, Section 1.1, (a) Annex VII, Section 4.10 Annex XIV, part B, Section 6.2 (a) | Obtain clinical literature relevant to the condition being treated with the device or obtain literature relevant to similar devices |
Systematic Literature Search as a source of clinical data under the MDR
In addition to clinical investigations and clinically relevant information from post-market surveillance, the scientific literature review is one of the few sources that can provide clinical data on a medical device (MDR, Article 2, (48)). Manufacturers can only consider the data obtained from the scientific literature as clinical data if this data comes “of a device for which equivalence to the device in question can be demonstrated”. This means that the first step when conducting a Systematic Literature Search is to identify similar devices and analyse the clinical, biological and technical characteristics to establish equivalence. Additional guidance on equivalence can be found here.
Once the equivalent devices have been established, the next steps can be conducted as described in the MEDDEV 2.7/1 Revision 4:
- Background to the literature search and the literature review: This section focuses on the literature search and review rationale. The following elements should be included: information on the device, link to risk management, comment on previous reviews, name, and model of equivalent devices.
- Objective: This section collects the research question and describes the search input; an example of search input can be the PICO methodology (P-Population, I-Intervention, C-Comparator, and O-Outcome). The search’s input should nourish from the device’s intended purpose, intended performance and clinical benefits, and risk management output.
- Methods: This section should contain the detailed protocol for the search execution, including sources of data, the extent of the search, exact search terms, data limits, inclusion/exclusion criteria, appraisal plan, and analysis plan.
- Appraisal of clinical data: The data collected through the search methodology is appraised for relevance. The appraisal should consider the quality of the studies obtained through the literature and discard those that lack scientific validity.
- Analysis of the clinical data: the data that fits the inclusion criteria should be analyzed for relevance against the device in question. The analysis must be objectively and critically evaluate for both favorable and unfavorable data. If the clinical data analysis reveals any gaps, a clinical investigation should be conducted to cover these gaps.
MEDDEV vs MDCGs guidance documents
Although several MDCG guidance documents have been published on the topic of clinical evaluation, none of them specifically describes how to conduct a systematic literature search and review. Thus, MEDDEV 2.7/1 Revision 4 continues to be the go-to reference document when conducting a literature review.
Despite this, some MDCG documents provide additional insights on the use of literature. MDCG “2020-1 Guidance on Clinical Evaluation (MDR) / Performance Evaluation (IVDR) of Medical Device Software” establishes that the steps for the clinical evaluation of Software as Medical Device (SaMD) are:
- Establish a valid clinical association
- Validate the technical performance
- Validate the clinical performance
MDCG2020-1 additionally mentions that literature searches can be used to support a valid clinical association between the output of a SaMD and the targeted clinical condition. This guidance also proposes that literature searches can be used to evaluate the clinical STATE-OF-THE-ART, including alternative diagnostic and treatment options. Even if equivalence cannot be established, literature searches can be used to establish a valid clinical association and perform the SOTA evaluation. This principle can also be applied to other low-risk devices in the absence of specific guidance.
In addition, “MDCG 2020-6 Regulation (EU) 2017/745: Clinical evidence needed for medical devices previously CE marked under Directives 93/42/EEC or 90/385/EEC” proposes to use systematic reviews of clinical data published in the literature to fill the gaps in clinical data. The use of high-quality studies is always preferred. However, this guidance document proposes that the lower-quality studies with “potential methodological flaws but where data can still be quantified” may be used as part of the literature review for legacy devices where no safety or performance concerns have been identified.
Common mistakes
One of the common mistakes that we find when examining the systematic scientific literature review documentation is the lack of a literature search methodology. Without a literature search protocol that clearly defines the selected keywords, the inclusion/exclusion criteria and the suitability criteria, it becomes challenging to replicate (and therefore assess) a literature search.
Another standard mistake is the use of literature from devices that are not demonstrated to be equivalent to the device under evaluation. The MEDDEV guidance and the MDR specifically mention that only the literature from equivalent devices may be used to obtain clinical data on the device under evaluation. Literature using other devices may be used on the CER but not as clinical evidence.
A third mistake, only applicable for medical devices that already have the CE-mark, is overlooking the PMS data. Manufacturers should proactively collect PMS data from their device as well as from competitor devices. Manufacturers can do this by searching for additional literature or looking into incident databases such as MAUDE.
Experts in Literature Search and Review
AKRN specializes in Systematic Literature search and review compliant with the Regulation (EU) 2017/745. As part of our medical writing services, AKRN can help medical device manufacturers fulfill the requirements needed to develop a methodology for a systematic literature search and obtain clinical evidence for the literature review for your new medical device.
We can also help your company update the literature review of a CE-marked or legacy device to an MDR compliant document. Our medical writers at AKRN can perform MDD to MDR gap assessments, provide guidance to complete a literature search and review transition from MDD or MEDDEV 2.7/1 revision 4 to the new MDR, and develop required documentation adapted to your needs.
