State of the Art (SOTA) Analysis under EU MDR 2017/745
The State of the Art analysis is the most misunderstood component of clinical evaluation under the MDR. It is not a literature search. It is not a summary of published papers. SOTA is the structured, clinically reasoned foundation that defines what your device must be measured against. and most manufacturers get it wrong before they even begin.
What Is State of the Art (SOTA) under the MDR?
State of the Art (SOTA) is the current body of knowledge available at the time of assessment, encompassing all that is known about a specific medical condition, its diagnosis, its treatment, and the technologies used to manage it. Under EU MDR 2017/745, the SOTA analysis is a mandatory element of the clinical evaluation process. It provides the clinical context against which your device’s performance and safety are judged.
The most critical misunderstanding in the industry is treating SOTA as a literature search. It is not. A literature search is a method for collecting data. SOTA is a clinical reasoning process that begins with understanding the disease or condition your device addresses, mapping the full treatment pathway, identifying all available alternatives. including surgical, pharmacological, and non-device options. and from this analysis, extracting the benefit-risk parameters and acceptance criteria that define what acceptable performance and safety look like for your device category.
SOTA is the lens through which every other component of your clinical evaluation is focused. It informs the acceptance criteria in your Clinical Evaluation Plan (CEP), drives the benefit-risk determination in your Clinical Evaluation Report (CER), and shapes the endpoints you monitor in your Post-Market Clinical Follow-up (PMCF) plan and report. Without a sound SOTA analysis, your entire clinical evaluation framework is built on an unstable foundation.
Why SOTA Begins with Clinical Understanding, Not Search Queries
A Notified Body reviewer does not open your SOTA section expecting a list of PubMed citations. They expect to see that you understand the clinical problem. What is the disease or condition? What symptoms drive patients to seek treatment? What is the natural history if left untreated? What are the established treatment pathways. first-line, second-line, surgical, conservative? Where does your device sit in that pathway?
Only after mapping this clinical landscape do you have the foundation to identify what performance benchmarks exist, what complication rates are considered acceptable, what clinical outcomes matter, and what safety thresholds have been established. These become your acceptance criteria. the quantified targets your device must meet or exceed to demonstrate an acceptable benefit-risk profile.
We addressed this exact methodology in depth during our SOTA Webinar with SZUTEST NB2975, where a Notified Body reviewer explained what they actually look for when they open the SOTA section of a clinical evaluation report. If you have not watched it, it is one of the most practical resources available on this topic.
What SOTA Defines
When performed correctly, the SOTA analysis delivers the following outputs:
- Clinical context: A complete description of the disease, condition, or clinical need, including epidemiology, pathophysiology, patient population characteristics, and disease staging or severity classifications.
- Treatment pathway mapping: A structured overview of all available treatment and management options. device-based, pharmacological, surgical, and conservative. along with their clinical positioning (first-line vs. last resort, curative vs. palliative).
- Benefit-risk parameters: The specific clinical endpoints, performance metrics, and safety metrics that are relevant to the device category, derived from clinical practice guidelines, meta-analyses, and consensus statements.
- Acceptance criteria: Quantified thresholds for each benefit-risk parameter, based on what the current state of the art considers acceptable performance and tolerable risk.
- Evidence base: The body of evidence from which these parameters and criteria are derived, including clinical guidelines, systematic reviews, registries, standards, and clinical investigation reports.
SOTA Analysis vs. Literature Review: They Are Not the Same
One of the most persistent errors in clinical evaluation practice is conflating the SOTA analysis with a literature review. While the literature review is a data-gathering activity, the SOTA analysis is a clinical reasoning exercise. The table below clarifies the fundamental differences:
| Dimension | SOTA Analysis | Literature Review |
|---|---|---|
| Purpose | Define the current state of knowledge for the clinical area, identify benefit-risk parameters and acceptance criteria | Systematically collect and appraise published evidence relevant to the subject device |
| Scope | Entire disease/condition landscape: all treatment options, clinical guidelines, consensus statements, non-device alternatives | Evidence specifically related to the subject device and/or equivalent devices |
| Output | Acceptance criteria, benchmark values, benefit-risk parameters, treatment pathway map | Appraised dataset of publications with quality scores, inclusion/exclusion rationale, data extraction tables |
| Methodology | Clinical analysis driven by disease understanding, guideline review, and treatment pathway mapping | Systematic search protocol with defined databases, search strings, PICO framework, and screening criteria |
| Sources | Clinical practice guidelines, consensus statements, HTAs, registries, standards, textbooks, clinical expert input | PubMed, Embase, Cochrane, CNKI, regulatory databases, clinical investigation reports |
| Update Trigger | New guidelines, new treatment alternatives, paradigm shifts in clinical practice, new safety signals | Periodic update per CEP schedule, new publications, PMCF data, vigilance findings |
When a manufacturer treats SOTA as simply “another literature search,” the result is typically a collection of abstracts with no clinical reasoning, no acceptance criteria, and no connection to the benefit-risk determination. This is one of the most frequently cited deficiencies in Notified Body review letters.
SOTA Methodology: A Step-by-Step Approach
The following six-step methodology reflects the approach outlined in MDCG 2020-6, MEDDEV 2.7/1 Rev. 4, and the expectations communicated by Notified Body reviewers. including the methodology we discussed in our SOTA Webinar with SZUTEST NB2975.
1. Define the Disease or Condition
Start with a thorough description of the medical condition your device addresses. Include epidemiology, pathophysiology, risk factors, patient demographics, disease classification systems, and staging criteria. This establishes the clinical context that frames everything that follows. Without understanding the condition, you cannot identify the right benchmarks.
2. Map the Treatment Pathway
Identify every available treatment option for the target condition. not just device-based alternatives. Include pharmacological treatments, surgical interventions, conservative management, physiotherapy, watchful waiting, and any other clinically recognized approach. Position each option in the treatment pathway: first-line, second-line, adjunct, or last resort. Explain where your device fits.
3. Identify All Alternatives (Including Non-Device)
Go beyond competitor devices. A Notified Body reviewer expects to see all alternatives a clinician would consider, not just the devices in your market segment. If the condition can be managed with medication, the SOTA must include that option and its outcomes. Missing non-device alternatives is a common deficiency that signals a superficial analysis.
4. Extract Benefit-Risk Parameters
From the treatment pathway analysis and clinical guidelines, extract the specific clinical endpoints that define benefit (e.g, procedure success rate, healing time, symptom resolution, functional improvement scores) and risk (e.g, complication rates, adverse event rates, re-intervention rates, mortality). These parameters must be clinically meaningful, measurable, and relevant to your device’s intended purpose.
5. Establish Acceptance Criteria
For each benefit-risk parameter, define a quantified acceptance criterion. What success rate does the literature support as acceptable? What complication rate is considered tolerable for this treatment category? These values come from clinical guidelines, meta-analyses, large registries, and systematic reviews. They become the benchmarks your device data will be measured against in the CER.
6. Document Evidence Sources
Record the sources used to derive each parameter and acceptance criterion. This includes clinical practice guidelines (with issuing body, year, and grade of recommendation), systematic reviews (with databases searched, number of studies included), registry data (with registry name, patient volume, follow-up period), and relevant standards. Traceability from criterion to source is essential for reviewer confidence.
Common SOTA Deficiencies Identified by Notified Bodies
Based on Notified Body feedback, MDCG 2020-13 assessment criteria, and insights shared during our SOTA Webinar with SZUTEST NB2975, these are the deficiencies that most frequently result in review queries or rejection:
Treating SOTA as a Literature Search
The most fundamental error. When the SOTA section reads like a list of abstracts rather than a clinical reasoning document, reviewers immediately flag it. SOTA must demonstrate understanding of the clinical landscape, not just that papers were found. It requires synthesis, not summarization.
No Acceptance Criteria Defined
A SOTA analysis without quantified acceptance criteria fails its primary purpose. The reviewer cannot assess whether your device data is adequate if no benchmarks have been established. Every benefit-risk parameter needs a clearly stated, evidence-based threshold value.
Only One Database Used
Relying solely on PubMed. or any single database. is insufficient for a robust SOTA. Clinical practice guidelines from specialty societies, health technology assessment reports, large device registries, and regional databases like CNKI each contribute different perspectives. A comprehensive SOTA draws from multiple, complementary sources.
Missing Non-Device Alternatives
If a patient’s condition can be managed with medication, surgery, physiotherapy, or conservative management, and these options are absent from your SOTA, the analysis is incomplete. Notified Body reviewers specifically check whether all clinically relevant alternatives are addressed, not just competing devices.
No Link to Risk Management
The SOTA analysis must connect to the risk management file. Hazards identified in the state of the art. known complications, adverse event profiles of alternative treatments, residual risks. should be reflected in the risk analysis (ISO 14971). A disconnected SOTA suggests siloed documentation rather than an integrated evaluation.
SOTA Not Updated with New Evidence
The state of the art evolves. New clinical guidelines are published, treatment paradigms shift, new safety signals emerge. If your SOTA was written three years ago and has not been updated to reflect new guidelines, registry data, or treatment alternatives, it no longer represents the current state of the art. MDR requires ongoing clinical evaluation, and the SOTA is part of that lifecycle.
How We Help with Your SOTA Analysis
Dr. Hatem Rabeh and the Clinical Evaluation Navigator team deliver SOTA analyses that satisfy Notified Body expectations. built on clinical reasoning, not template filling:
- ✓ SOTA Report Writing Complete State of the Art analysis from disease definition through acceptance criteria derivation, structured to meet MDCG 2020-6 and MDCG 2020-13 expectations, ready for integration into your CER
- ✓ Treatment Pathway Mapping Clinically accurate mapping of the full treatment landscape, including device-based, pharmacological, surgical, and conservative management options, positioned within the clinical decision pathway
- ✓ Benefit-Risk Parameter Definition Identification and extraction of clinically meaningful benefit and risk endpoints from clinical practice guidelines, systematic reviews, meta-analyses, and large device registries
- ✓ Acceptance Criteria Derivation Quantified, evidence-based acceptance thresholds for each benefit-risk parameter, with full traceability to source documents and grade of evidence
- ✓ Database Strategy & Source Documentation Multi-database sourcing strategy covering clinical guideline repositories, PubMed, Embase, Cochrane, HTAs, NICE, specialty society guidelines, and registries relevant to your device’s clinical area
- ✓ Notified Body Response Support Drafting targeted responses to NB deficiency letters related to SOTA, including supplemental analyses, additional guideline references, and revised acceptance criteria documentation
Frequently Asked Questions
What is State of the Art (SOTA) under the MDR?
State of the Art under EU MDR 2017/745 refers to the current body of technical and clinical knowledge available at the time of assessment. In the context of clinical evaluation, it encompasses everything that is known about a medical condition, its available treatments (device-based and non-device), accepted clinical outcomes, established benefit-risk parameters, and quantified acceptance criteria. SOTA is defined in MDR Annex XIV Part A and is referenced in MDCG 2020-6 and MDCG 2020-13. It is not a literature search. it is a clinical reasoning process that provides the benchmark framework against which your device’s performance and safety data are evaluated.
Is SOTA the same as a literature review?
No. This is the single most common misconception in clinical evaluation practice. A literature review is a systematic data-gathering activity that collects and appraises published evidence about a specific device or technology. The SOTA analysis is a clinical reasoning exercise that defines the entire landscape of knowledge about a condition, its treatments, and the established benchmarks for acceptable performance and safety. The literature review feeds data into the CER. The SOTA analysis defines what that data must be measured against. They serve fundamentally different purposes, use different methodologies, and produce different outputs.
Which databases should I use for SOTA?
SOTA draws from a broader range of sources than a standard literature search. Key sources include: Clinical practice guidelines from specialty societies (e.g, ESC, AHA, EAU, NICE), systematic reviews and meta-analyses from the Cochrane Library, health technology assessments (HTA bodies like NICE, G-BA, HAS), large device registries (e.g, NJR, EACTS, ELSO), consensus statements, and relevant harmonized standards. PubMed and Embase are supplementary for identifying primary studies, but the SOTA should prioritize synthesized, high-level evidence sources that reflect current clinical consensus.
Does SOTA apply to Class I devices under Article 61(10)?
Yes. Article 61(10) provides certain exemptions for well-established Class I and specific Class IIa devices regarding the depth of clinical data required, but it does not exempt them from clinical evaluation or from SOTA analysis. Even for devices that qualify under Article 61(10), the manufacturer must demonstrate that they understand the clinical context, have identified relevant benefit-risk parameters from the state of the art, and can justify their device’s safety and performance claims. The scope of the SOTA may be narrower for simpler devices, but the obligation to perform it remains.
How often should SOTA be updated?
The SOTA analysis must be updated whenever there is a material change in the state of the art. for example, when new clinical practice guidelines are published, when new treatment alternatives emerge, when safety signals change the accepted risk profile, or when significant new registry data becomes available. In practice, the SOTA should be reviewed at least as frequently as the CER update cycle: annually for Class III and implantable devices, and at defined intervals for other classes as specified in the Clinical Evaluation Plan. If no changes have occurred, a brief documented justification is sufficient, but the review itself must still happen.
Watch Our SOTA Webinar with SZUTEST NB2975
In one of the most practical educational sessions available on SOTA methodology, Dr. Hatem Rabeh hosted a live webinar with a reviewer from SZUTEST, Notified Body 2975, focused entirely on the State of the Art analysis under the MDR. The webinar addressed the exact questions manufacturers struggle with most:
- What do Notified Body reviewers actually look for when they open the SOTA section of a CER?
- How should SOTA be structured to demonstrate clinical reasoning rather than just literature listing?
- What are the most common SOTA deficiencies that trigger review queries?
- How should acceptance criteria be derived and presented?
- What level of detail is expected for treatment pathway mapping?
This webinar has become one of the most referenced resources in the Clinical Evaluation Navigator community, precisely because it provides direct Notified Body perspective on a topic where manufacturer practice often diverges significantly from reviewer expectations.
Related Articles
Need Help with Your SOTA Analysis?
Join the Clinical Evaluation Navigator community for expert guidance on SOTA analysis, benefit-risk parameter definition, acceptance criteria derivation, and the full clinical evaluation process under the MDR. Get direct access to Dr. Hatem Rabeh, ask your regulatory questions, and learn from real Notified Body feedback.
Free to join • Weekly expert calls • SOTA webinar archive access
