How to Conduct an EU MDR Literature Review: A Step-by-Step Guide to Submission-Ready Evidence

Applying PRISMA Principles to Meet MEDDEV 2.7/1 Rev 4 and Annex XIV Requirements

Starting an EU MDR literature review can feel overwhelming. Annex XIV outlines what manufacturers must achieve, but provides little guidance on how to build a process that holds up to Notified Body scrutiny.

Regulatory expectations around transparency, reproducibility, and methodological defensibility continue to increase. Reviewers assess not only what evidence was found, but how it was searched, screened, evaluated, interpreted, and integrated into clinical evaluation activities.

When a literature review lacks clear methodology, traceable screening decisions, or alignment with clinical evaluation objectives, it weakens the defensibility of the clinical evidence package, even when the team has identified large volumes of literature. Notified Bodies now expect structured workflows, consistent documentation, and clear reviewer accountability throughout every stage of the MDR literature review process.

Developing a submission-ready EU MDR literature review requires more than performing database searches. It means establishing a structured and reproducible process that supports evidence traceability from protocol development through to final reporting.

This guide walks through the MDR systematic literature review process step by step, covering protocol development, database searching, evidence screening, critical appraisal, PRISMA documentation, and preparation of a submission-ready literature review report.

What the MDR Requires From Your Literature Review

Under EU MDR 2017/745, literature reviews must follow a systematic, objective, and reproducible methodology that supports the clinical evaluation process. Annex XIV Part A requires manufacturers to identify, appraise, and analyse all relevant clinical data to demonstrate device safety, clinical performance, clinical benefit, and benefit-risk profiles in support of regulatory compliance.

A systematic literature review is a critical to achieving compliance and of supporting regulatory submissions and technical documentation.

MEDDEV 2.7/1 Rev 4 remains a key methodological reference for literature review and clinical evaluation activities. Notified Bodies assess whether search strategies, screening decisions, evidence appraisal methods, and documentation practices align with established clinical evaluation principles. Manufacturers also struggle to determine the amount of clinical evidence needed for compliance, especially because the MDR removes grandfathering and all devices require recertification, which contributes to long approval timelines.

Step 1: Write Your Search Protocol

Before performing any database searches, establish a literature search protocol that defines how evidence will be identified, screened, evaluated, and documented to improve Reproducibility, reduce reviewer bias, and demonstrate methodological consistency during Notified Body review.

Begin by clarifying the device scope including the subject device, intended purpose, target patient population, indications for use, risk profile, the relevant medical field, and the clinical background needed to define the state of the art. Defining these elements early determines which evidence sources are clinically relevant and which fall out of scope.

State-of-the-art searches establish the current clinical and technological landscape, including generally accepted treatment options and standards. SOTA reviews are essential for compliance with EU MDR and IVDR. Equivalence searches evaluate evidence from comparable devices that may support the device under evaluation. Searches may also assess an equivalent device and similar competitor devices when relevant to the clinical evaluation. This includes understanding current treatment approaches, comparable technologies, clinical practice standards, and competing therapeutic options relevant to the intended purpose.

Many MDR literature review workflows incorporate the PICO framework to structure clinical questions by defining the target population, intervention or device, comparator context, and intended outcomes. This approach aligns with expectations outlined in MEDDEV 2.7/1 Rev 4. Predefined research questions should guide the search and evidence synthesis, and from these questions, teams should establish predefined inclusion and exclusion criteria covering patient populations, study designs, publication dates, language, device relevance, comparator relevance, and clinical outcomes. Defining these criteria before searching begins reduces subjective screening decisions, improves reviewer consistency, and strengthens audit readiness.

Review objectives should reflect the specific evidence requirements of the clinical evaluation. Some reviews focus on demonstrating safety and performance, while others emphasise benefit-risk evaluation, long-term outcomes, PMCF integration, or specific concerns identified through post-market surveillance activities. Objectives may also include demonstrating clinical benefit, maintaining benefit-risk profiles, and integrating new clinical data over the device lifecycle. Post-market objectives should also align with risk management; a PMS system must be maintained for each medical device, and PMCF is required for all Class IIb and Class III medical devices.

Step 2: Run Your Database Searches

Perform structured database searches to identify potentially relevant evidence from medical literature, scientific literature, and published literature. Select databases that align with the device type, intended purpose, therapeutic area, and clinical questions, and that capture all relevant data. Commonly used databases include PubMed, Embase, and the Cochrane Library, along with specialty-specific databases relevant to the device.

Search strategies combine predefined keywords, subject headings, device terminology, clinical conditions, synonyms, and outcome-related terms using Boolean operators.

EU MDR literature reviews often require supplementary evidence sources including conference abstracts, clinical trial registries, grey literature, vigilance databases, safety communications, PMCF findings, and manufacturer complaint data. Performance evaluation for an in vitro diagnostic may also require targeted searches for scientific validity, clinical performance, and evidence used in performance evaluation reports.

Document databases searched, search dates, complete search strings, applied filters, search limitations, and the rationale for evidence source selection. Search documentation must be sufficiently detailed to allow another reviewer to reproduce the search independently, while also supporting regulatory submissions and inclusion in technical documentation.

Step 3: Deduplicate and Import

Import retrieved records into a literature review platform or reference management system. Records from multiple databases are consolidated into a single dataset before screening begins.

Because the same publication may be indexed across multiple databases, duplicate records must be identified and removed before screening. Accurate duplicate management maintains reliable PRISMA counts and ensures transparency throughout the review process. Document the number of duplicate records removed and the methodology used to identify them.

Step 4: Title and Abstract Screening

Screen titles and abstracts against the predefined inclusion and exclusion criteria established in the review protocol. Publications clearly irrelevant to the intended purpose, device scope, target population, or clinical objectives are excluded before progressing to full-text assessment.

Consistent application of inclusion and exclusion criteria is critical under EU MDR. Inconsistency introduces selection bias, reduces reproducibility, and weakens the overall defensibility of the evidence review. For a detailed overview of the most common literature review challenges that arise during screening and appraisal, see our dedicated article on this topic.

Step 5: Full-Text Review

Studies passing title and abstract screening proceed to full-text review. Assess each publication against the predefined clinical and methodological requirements established in the review protocol, determining whether it provides clinically relevant data for the subject device or an equivalent device, including any clinical investigation evidence.

Extract key study information using a structured data extraction form covering study characteristics, patient populations, device details, comparator information, safety outcomes, clinical performance, performance data, adverse events, follow-up duration, and reviewer comments.

Perform critical appraisal to assess the quality and relevance of evidence, including evaluation of study design, methodological limitations, risk of bias, endpoint selection, statistical robustness, patient population relevance, and overall applicability to the intended purpose of the device. MEDDEV 2.7/1 Rev 4 provides detailed guidance on appraisal and weighting methodologies that align with Notified Body expectations.

Interpret evidence within the broader context, considering current treatment standards, clinical practice guidelines, accepted standards within the medical field, comparable technologies, and competing therapeutic approaches. Article 61 of EU MDR, requires clinical evidence to be evaluated in the context of currently available alternatives.

Maintain a clear audit trail throughout. Document extraction decisions, appraisal outcomes, reviewer comments, and exclusion rationale to strengthen traceability and support alignment between evidence appraisal and final CER conclusions.

Step 6: Generate Your PRISMA Flow Diagram

Document the study selection process using a PRISMA flow diagram. Under EU MDR, PRISMA-style documentation demonstrates methodological rigour and provides a clear audit trail of how evidence was identified and selected.

The PRISMA flow diagram captures key metrics from each stage of the review:

• Records identified through database searching
• Records identified through additional evidence sources
• Duplicate records removed
• Records screened at title and abstract stage
• Records excluded during screening
• Full-text articles assessed for eligibility
• Full-text articles excluded with documented reasons
• Studies included in the final literature review

Numbers reported in the PRISMA diagram must align with search outputs, screening logs, exclusion records, and final evidence tables.

Step 7: Write the Literature Report

Develop a structured, submission-ready literature review report that supports regulatory compliance and submissions by clearly demonstrating how evidence was identified, screened, appraised, interpreted, and incorporated into clinical conclusions supporting the device’s safety, performance, and benefit-risk profile.

Present evidence objectively. EU MDR 2017/745 requires manufacturers to consider both favourable and unfavourable data when generating clinical evidence.

Link evidence findings clearly to final clinical conclusions. Review outputs should support the conclusions presented in the Clinical Evaluation Report and lifecycle evidence activities including PMS assessments, PMCF planning, state-of-the-art evaluations, and ongoing benefit-risk monitoring. They should also support technical documentation, performance evaluation reports where applicable, and ongoing risk management.

Common Mistakes That Fail Notified Body Review

Common literature review deficiencies identified during Notified Body review, and weak literature review processes can delay achieving compliance under MDR, relate to methodology transparency, documentation consistency, and workflow traceability rather than the absence of evidence itself.

For a full breakdown of the most common reasons Clinical Evaluation Reports are rejected during Notified Body review, see our dedicated article on this topic.

Building a Defensible Literature Review: Why Every Stage Matters

From protocol development and database searching through to screening, critical appraisal, PRISMA documentation, and final reporting, each stage of an EU MDR literature review contributes to the quality and defensibility of the clinical evidence package. Notified Bodies expect reviews to be transparent, reproducible, and supported by clear documentation in line with MEDDEV 2.7/1 Rev 4 and EU MDR 2017/745. A well-executed literature review establishes a reliable foundation for CER development, PMS activities, PMCF planning, and ongoing clinical evidence generation throughout the device lifecycle. Stringent MDR requirements, recertification of legacy devices, and long approval timelines make a proactive documentation approach especially important.

If you are preparing a literature review for a Clinical Evaluation Report or post-market surveillance update and want support building a structured, audit-ready process, Citemeds can help. Our team works with medical device manufacturers to design and conduct literature reviews that meet EU MDR expectations from protocol development through to final reporting, including support for technical documentation and regulatory submissions. Get in touch to discuss your requirements.

Frequently Asked Questions

What does an EU MDR literature review need to include?

An EU MDR literature review must include a documented search protocol defining the review objective, databases, search terms, and inclusion and exclusion criteria; structured database searches across sources such as PubMed, Embase, and the Cochrane Library; systematic title and abstract screening with documented decisions; full-text review with critical appraisal of included studies; PRISMA flow diagram documentation; and a structured literature review report linking evidence findings to clinical evaluation conclusions. The process should also be driven by predefined research questions and a literature search protocol. Each stage must be transparent, reproducible, and traceable to support Notified Body assessment under EU MDR 2017/745.

What databases should be searched for an EU MDR literature review?

The minimum database set typically includes PubMed, Embase, and the Cochrane Library. Depending on the device type and therapeutic area, additional specialty-specific databases may also be required. Supplementary evidence sources such as ClinicalTrials.gov, grey literature, vigilance databases, and PMCF findings should also be considered. The rationale for database selection should be documented within the search protocol and sufficiently detailed to allow another reviewer to reproduce it independently. Database selection should be sufficient to capture all data relevant and clinical data relevant to the review objectives.

What is a PRISMA flow diagram and is it required under EU MDR?

A PRISMA flow diagram is a visual summary of how records progressed through each stage of the literature review, from initial identification through to final inclusion. While EU MDR 2017/745 does not explicitly mandate PRISMA, Notified Bodies widely expect it as a standard for demonstrating transparency and reproducibility. The diagram should accurately reflect the number of records identified, deduplicated, screened, excluded, and included, with all counts consistent across search outputs, screening logs, and evidence tables.

What are the most common reasons EU MDR literature reviews fail Notified Body review?

The most common reasons include undocumented exclusion decisions, missing or poorly justified database selection, inconsistent PRISMA counts, weak search documentation, inconsistent application of inclusion and exclusion criteria, inadequate state of the art assessment, lack of clinical background depth, weak linkage to risk management, and insufficient linkage between evidence findings and clinical conclusions. For a full breakdown see our article on reasons Clinical Evaluation Reports are rejected.

What is critical appraisal in a medical device literature review?

Critical appraisal is the structured process of evaluating the quality, relevance, and methodological robustness of studies included in a literature review. In a medical device context this includes assessing study design, risk of bias, endpoint selection, statistical robustness, patient population relevance, follow-up duration, and overall applicability to the device’s intended use. MEDDEV 2.7/1 Rev 4 provides the primary methodological framework for appraisal in EU MDR clinical evaluation. Critical appraisal is one of the most frequently scrutinised aspects of literature review methodology during Notified Body assessment.