EU MDR Device Classification — A Practical Guide
Introduction
Before you can apply for CE marking, select a Notified Body, or build your Technical Documentation, you need to answer one foundational question: what class is your device?
Device classification under EU MDR is not just an administrative formality. It determines your conformity assessment route, the level of clinical evidence required, your post-market surveillance obligations, and how much Notified Body involvement your submission will need. Get it wrong — even by one class — and you could be building your entire regulatory strategy on the wrong foundation.
The good news is that the classification system, while detailed, follows a logical structure. EU MDR Annex VIII defines 22 classification rules covering non-invasive devices, invasive devices, active devices, and special rules for specific device categories. Once you understand the logic behind the rules, classification becomes a systematic process rather than a guessing game.
This guide walks you through the EU MDR classification framework, explains each device class and its implications, and gives you a practical step-by-step approach to classifying your device correctly.
The Four Device Classes
EU MDR divides medical devices into four classes based on risk — Class I, Class IIa, Class IIb, and Class III. The higher the class, the higher the risk, and the more rigorous the regulatory pathway.
Class I — Low Risk
Class I devices are considered low risk and are subject to General Controls only. Most Class I devices can self-declare conformity through a Declaration of Conformity without involving a Notified Body — making this the most straightforward pathway to CE marking.
However, there are important exceptions. Class I devices that are supplied in sterile condition, have a measuring function, or are reusable surgical instruments require Notified Body involvement for those specific aspects — even though the overall device remains Class I. These are sometimes referred to as Class Is (sterile), Class Im (measuring), and Class Ir (reusable surgical).
Examples of Class I devices include non-invasive devices such as bandages, wound dressings, and hospital beds, as well as corrective spectacles, stethoscopes, and manual wheelchairs.
Class IIa — Medium Risk
Class IIa devices carry moderate risk and require Notified Body involvement for conformity assessment. The NB will assess either your Quality Management System under Annex IX, or conduct a type examination under Annex XI.
Examples include hearing aids, surgical gloves, short-term invasive catheters, ultrasound imaging systems, and dental fillings.
Class IIb — Medium-High Risk
Class IIb devices carry a higher level of risk and require a more rigorous conformity assessment. The NB must assess your QMS under Annex IX plus conduct a type examination under Annex X, or carry out product verification under Annex XI.
Examples include ventilators, infusion pumps, long-term surgical implants, bone fixation plates, and intensive care monitoring equipment.
Class III — High Risk
Class III is reserved for the highest-risk devices — those that are implantable, life-sustaining, or present the greatest potential for harm. The conformity assessment is the most demanding, requiring full QMS scrutiny under Annex IX combined with mandatory product scrutiny under Annex X.
For certain Class III devices, MDR also requires a clinical evaluation consultation procedure involving an expert panel — an additional layer of scientific scrutiny introduced by MDR that did not exist under MDD.
Examples include drug-eluting stents, pacemakers, total joint replacements, spinal implants, and implantable defibrillators.
How the Classification Rules Work
MDR Annex VIII contains 22 classification rules organized into four groups. The rules are applied based on the nature of the device, how it interacts with the body, its duration of contact, and its intended purpose.
The four groups are non-invasive devices (Rules 1–4), invasive devices (Rules 5–8), active devices (Rules 9–13), and special rules for specific device types (Rules 14–22).
The key principle is that the highest applicable rule determines the final class. If your device falls under multiple rules, you apply each one and assign the class corresponding to the highest result.
Non-Invasive Devices — Rules 1 to 4
Rule 1 covers all non-invasive devices not addressed by any other rule. These are Class I by default — the lowest risk category.
Rule 2 applies to non-invasive devices intended to channel or store blood, body fluids, cells, or tissues. Most of these are Class IIa, but devices used for storing or channelling blood or other body fluids for eventual infusion into the body may be Class IIb if the interaction with the body is significant.
Rule 3 covers non-invasive devices that modify the biological or chemical composition of blood, other body fluids, or cells intended for transfusion or administration. These are generally Class IIb, or Class III if the treatment involves extracorporeal circulation.
Rule 4 applies to non-invasive devices that come into contact with injured skin or mucous membranes. Wound dressings for superficial wounds are Class I, while devices managing the micro-environment of a wound are Class IIa or IIb depending on the wound type.
Invasive Devices — Rules 5 to 8
Invasive devices — those that penetrate the body through a body orifice or through the surface of the body — are governed by Rules 5 through 8, with classification depending primarily on the duration and site of contact.
Rule 5 covers invasive devices used in body orifices other than surgically created ones. Transient use (less than 60 minutes) is Class I. Short-term use (up to 30 days) connected to an active device of Class IIa or higher is Class IIa. Long-term use (more than 30 days) is Class IIb.
Rule 6 applies to surgically invasive devices for transient use. Most are Class IIa, but those specifically intended to diagnose, monitor, or correct a cardiac defect are Class III.
Rule 7 covers surgically invasive devices for short-term use — up to 30 days. These are generally Class IIa, with Class IIb or III applying to devices that undergo chemical change in the body or are intended to administer medicines.
Rule 8 is one of the most significant rules — it covers implantable devices and long-term surgically invasive devices. The default class is Class IIb. However, devices that are intended to be placed in the heart, central circulatory system, or central nervous system are Class III. Devices that come into contact with the spinal column or that are intended to have a biological effect are also Class III.
Active Devices — Rules 9 to 13
Active devices — those that depend on a source of energy other than that generated by the human body — are covered by Rules 9 through 13.
Rule 9 covers active therapeutic devices. Devices intended to administer or exchange energy are Class IIa. Those intended to control or monitor vital physiological processes are Class IIb. Devices that emit ionizing radiation or are intended to administer radiopharmaceuticals are Class IIb or Class III.
Rule 10 applies to active devices for diagnosis and monitoring. Devices intended to supply information through in vitro examination of specimens derived from the human body are covered by the IVDR, not MDR. Diagnostic imaging devices using ionizing radiation are Class IIb. Devices that monitor vital physiological parameters where variation could result in immediate danger are Class IIb.
Rule 11 is particularly important in 2025 because it covers software. Software intended to provide information used to make decisions with diagnosis or therapeutic purposes is Class IIa at minimum. Software intended to monitor physiological processes is Class IIb. Software intended to take decisions that could cause death or irreversible deterioration of health is Class III. This rule is the primary reason why many Software as a Medical Device (SaMD) products are classified higher under MDR than they were under MDD.
Rule 12 covers active devices intended to administer medicines, body liquids, or other substances to or from the body. These are Class IIa as a minimum, with Class IIb or III applying if the substances involved are hazardous or if the device is life-sustaining.
Rule 13 applies to all other active devices not covered by the previous rules. The default is Class I.
Special Rules — Rules 14 to 22
Rules 14 through 22 address specific device categories where the general rules would result in an inappropriate classification. These special rules override the general rules when applicable.
Rule 14 covers devices incorporating a medicinal substance as an integral part. If the medicinal substance has action ancillary to the device, the device is Class III.
Rule 15 applies to devices used for contraception or prevention of sexually transmitted diseases. Most are Class IIb, with Class III for implantable or long-term invasive devices.
Rule 16 covers devices specifically intended for disinfecting, cleaning, rinsing, or hydrating contact lenses. These are Class IIb.
Rule 17 applies to devices specifically intended for recording diagnostic images generated by X-ray radiation — Class IIa.
Rule 18 covers devices using non-viable animal or human tissues or cells, or their derivatives. Devices incorporating non-viable animal tissues or cells rendered non-viable are Class III unless only used in external contact with intact skin.
Rule 19 applies to devices incorporating nanomaterials. Depending on the potential for internal exposure, these can be Class IIa, IIb, or III.
Rule 20 covers invasive devices with respect to body orifices that administer medicinal products by inhalation — Class IIa or IIb depending on the device.
Rule 21 applies to devices that are composed of substances or combinations of substances intended to be introduced into the human body via a body orifice or applied to the skin — classified by absorption and local or systemic effect.
Rule 22 covers active therapeutic devices with an integrated or incorporated diagnostic function which significantly determines the patient management — Class III.
