MDR classes explained for EU medical device classification

What MDR classes mean

MDR classes are the EU risk categories for medical devices under Regulation (EU) 2017/745. They do not apply to in vitro diagnostic devices, which are regulated separately under the IVDR. Article 51 and Annex VIII of the MDR divide devices into Class I, Class IIa, Class IIb and Class III based on intended purpose and inherent risks. Class I is generally the lowest-risk group, while Class III is the highest-risk group.

Class I can also be divided into Is for sterile devices, Im for devices with a measuring function, and Ir for reusable surgical instruments. The class is not a marketing preference or a product family label. It determines the conformity assessment route, the level of notified body involvement, clinical evidence expectations, post-market surveillance duties, and the parts of the technical documentation that need closer attention.

This overview reflects Regulation (EU) 2017/745 and the European Commission guidance MDCG 2021-24 Rev.1, Guidance on classification of medical devices, which the Commission lists with an April 2026 publication date. It is a regulatory explainer, not a final classification decision for any individual device.

MDR class	Risk position	Typical regulatory effect	Important caution
Class I	Lower risk	The manufacturer usually declares conformity after preparing technical documentation and meeting MDR requirements.	Class I is still regulated. Is, Im and Ir require notified body involvement for the relevant sterile, measuring or reusable surgical aspects.
Class IIa	Low to medium risk	Notified body assessment is generally required, and the periodic safety update report is updated at least every two years.	A simple product type label is not enough; intended use and applicable rules can move a device higher.
Class IIb	Medium to higher risk	Notified body involvement is more extensive, and the periodic safety update report is updated at least annually.	Implantable IIb devices and certain active therapeutic devices can trigger additional review expectations.
Class III	Highest risk	The most demanding conformity assessment, clinical evidence and post-market requirements apply.	Many, but not all, high-risk implants and devices with medicinal, biological or critical system effects fall here.

How Annex VIII assigns a device to an MDR class

Annex VIII is the classification engine of the MDR. It contains 22 rules that manufacturers must apply to the device as it is actually intended to be used. The analysis starts with the manufacturer’s intended medical purpose, because the same material, software module or instrument may fall into a different class if the claims, users, duration of use or patient population change.

The key principle is straightforward: if several rules or sub-rules apply, the strictest rule leading to the higher class controls the outcome. For that reason, classification should be documented as a reasoned analysis rather than a one-line conclusion. A technical file should show which rules were considered, which rule determined the class, and why competing rules were not applied.

The four rule families

Rule group	Scope	Classification drivers
Rules 1 to 4	Non-invasive devices	Contact with intact or injured skin, channelling or storing substances, modifying biological or chemical composition, and wound management purpose.
Rules 5 to 8	Invasive and surgically invasive devices	Body orifice use, surgical invasiveness, transient, short-term or long-term duration, implantable use, and contact with the heart, central circulatory system or central nervous system.
Rules 9 to 13	Active devices and software	Energy delivery, diagnostic or therapeutic function, administration or removal of substances, software decision impact and monitoring of physiological processes.
Rules 14 to 22	Special rules	Ancillary medicinal substances, contraception or prevention of sexually transmitted disease, device disinfection or sterilisation, diagnostic X-ray image recording, tissues or cells, nanomaterials, inhalation products, substance-based devices and certain closed-loop active therapeutic systems.

Inputs that should be fixed before classification

• The exact intended purpose, including indications, contraindications and clinical claims.
• The user group, such as lay users, healthcare professionals or trained specialists.
• The patient population and use environment, including home, hospital, emergency or intensive care settings.
• Whether the device is invasive, surgically invasive, implantable, active, reusable or sterile.
• The duration of continuous use, commonly assessed as transient, short-term or long-term.
• Whether the device incorporates software, a medicinal substance, human or animal material, nanomaterial or an absorbable substance.
• Whether accessories or components need separate classification in their own right.

Class I is not the same as no regulatory burden

A common misunderstanding is that Class I means little documentation. Under the MDR, Class I devices still need to comply with the general safety and performance requirements in Annex I, have technical documentation under Annex II, have post-market surveillance documentation under Annex III, carry a CE mark when applicable, and be supported by an EU declaration of conformity. Manufacturers must also address vigilance, UDI and registration obligations where applicable.

The main difference is the conformity assessment route. For ordinary Class I devices that are not sterile, do not have a measuring function and are not reusable surgical instruments, the manufacturer can generally declare conformity without a notified body certificate. For Class Is, Class Im and Class Ir, notified body involvement is limited to the relevant feature: sterile condition, measuring function or reusable surgical instrument aspects such as cleaning, disinfection, sterilisation, maintenance and functional testing.

Class I should therefore be treated as a documented regulatory conclusion, not a shortcut. If a non-invasive product touches injured skin, channels fluids, includes measurement claims, contains software or is supplied sterile, the classification rationale should address those features directly. Strong classification files explain not only why one rule applies, but also why higher-risk rules do not.

What changes as MDR classes increase

Higher MDR classes usually mean deeper third-party review, stronger clinical evidence expectations and more structured post-market obligations. The MDR sets baseline responsibilities for every class, but regulatory scrutiny increases as the potential harm to patients and users increases.

Class IIa

Class IIa covers many devices that are not high risk but go beyond ordinary Class I controls. Examples can include certain short-term invasive devices, active diagnostic devices and devices that manage the micro-environment of injured skin, depending on the exact intended purpose. Notified body assessment is generally required. Post-market surveillance includes a periodic safety update report that must be updated at least every two years.

Class IIb

Class IIb typically applies where a device presents a higher level of risk because of long-term invasiveness, hazardous administration or removal of substances, significant energy delivery, or higher-risk active therapeutic functions. Class IIb devices require notified body involvement and a periodic safety update report updated at least annually. Some IIb categories, such as certain implantable devices or active devices intended to administer or remove medicinal products, may attract additional documentation or clinical evaluation scrutiny under the MDR.

Class III

Class III is reserved for the highest-risk devices. It can include devices that come into direct contact with the heart, central circulatory system or central nervous system, certain implants, devices incorporating an ancillary medicinal substance, high-risk nanomaterial devices, and substance-based devices that are systemically absorbed or achieve their intended purpose in particularly critical ways. Class III classification affects technical documentation review, clinical evaluation, post-market follow-up and, for many devices, the summary of safety and clinical performance.

Software, substances and borderline products need extra care

Many classification errors occur when teams rely on old assumptions from the Medical Devices Directive or compare the product too broadly with similar devices. The MDR introduced or strengthened rules that are especially important for software, substance-based devices and special technologies. See also: Implants.

Software should be assessed under Rule 11 when it is intended to provide information used for diagnostic or therapeutic decisions, or to monitor physiological processes. The resulting class depends on the significance of the information and the seriousness of the healthcare situation. Software that merely stores, archives, communicates or displays data may fall outside the medical device software classification logic, but that conclusion depends on whether it performs analysis, drives treatment, changes dosage or directly supports a medical function.

Substance-based products need the same level of care. Rule 21 can apply to devices composed of substances or combinations of substances introduced into the human body through a body orifice or applied to the skin and absorbed by, or locally dispersed in, the body. If the principal intended action is pharmacological, immunological or metabolic, the product may instead raise medicinal product borderline questions rather than a straightforward MDR class question.

IVD products are another frequent source of confusion. MDR Classes I, IIa, IIb and III are not used for in vitro diagnostic medical devices under Regulation (EU) 2017/746. IVDs use IVDR Classes A, B, C and D, with a different Annex VIII rule set.

A practical workflow for classifying a medical device

A defensible classification process is structured, repeatable and connected to the technical documentation. The following workflow is useful for regulatory teams preparing a classification rationale.

1. Confirm the product is a medical device or an accessory under the MDR before assigning any class.
2. Freeze the intended purpose, claims, user profile, patient population and use environment.
3. Identify all relevant physical and functional characteristics, including invasiveness, activity, sterility, measuring function, duration of use and reusability.
4. Check whether the device includes software, medicinal substances, tissues or cells, nanomaterials, absorbable substances or other special-rule triggers.
5. Apply every potentially relevant Annex VIII rule, not only the rule that appears most convenient.
6. Use the strictest applicable rule or sub-rule when more than one rule applies.
7. Document excluded rules with short reasons, especially where a higher class could appear plausible.
8. Align the conformity assessment plan, clinical evaluation plan, labelling, UDI strategy and post-market surveillance plan with the final class.

If classification becomes disputed once a notified body is involved, the MDR provides mechanisms for referral to competent authorities. In practice, teams should try to avoid late disputes by preparing a clear rule-by-rule rationale before certification planning, budgeting and clinical evidence strategy are fixed. For more regulatory explainers, visit the Regulatory section.

Common mistakes when interpreting MDR classes

• Using device type instead of intended purpose. A dressing, catheter, app or instrument can move class depending on its claims and clinical use.
• Stopping after the first matching rule. Annex VIII requires consideration of all applicable rules, with the higher classification controlling where rules overlap.
• Treating Class I as a light documentation category. Class I devices still need MDR technical documentation, risk management, clinical evaluation, post-market surveillance and vigilance processes.
• Ignoring Class I subclasses. Sterile supply, measuring function or reusable surgical use can bring notified body involvement even when the base classification is Class I.
• Applying FDA classes to EU MDR decisions. US Class I, II and III categories are not interchangeable with EU MDR Classes I, IIa, IIb and III.
• Assuming old MDD classification still applies. MDR Annex VIII includes rules and interpretations that can up-classify certain software, substance-based, nanomaterial or special-purpose devices.
• Forgetting accessories. Accessories are classified in their own right, separately from the device with which they are used.

Frequently asked questions

What are the four MDR classes?

The four main MDR classes are Class I, Class IIa, Class IIb and Class III. They represent increasing levels of risk and regulatory scrutiny. Class I also has the subclasses Is, Im and Ir for sterile devices, measuring devices and reusable surgical instruments.

Who decides the MDR class of a device?

The manufacturer is responsible for classifying the device under Annex VIII and documenting the rationale in the technical documentation. For classes requiring notified body involvement, the notified body will review relevant aspects during conformity assessment. Competent authorities may become involved in classification disputes or complex borderline cases.

Does every Class IIa, IIb or III device need a notified body?

In general, yes. MDR conformity assessment for Class IIa, IIb and III devices involves a notified body. The depth of review differs by class, device type and chosen conformity assessment route.

Are IVD products covered by MDR classes?

No. In vitro diagnostic medical devices are regulated under the IVDR, not the MDR, and use Classes A, B, C and D. Applying MDR Classes I, IIa, IIb and III to IVDs is a common classification error.

Can a device class change after launch?

Yes. A class may need reassessment if the intended purpose, clinical claims, design, software function, material composition, duration of use or regulatory interpretation changes. Classification should be reviewed during change control and when new guidance affects the device category.