MDR Rule 19: Why a Nanomaterial Device Can Never Be Class I

EU MDR Annex VIII, Chapter III added four special rules the old directive never had, and Rule 19 is the one materials teams miss because it isn't a device-type rule at all — it's a materials rule. It doesn't ask what your device does. It asks what it's made of. And once the answer includes nanomaterial, one classification outcome disappears completely: Class I is gone, full stop, regardless of how low-risk everything else about the device looks.

What counts as a nanomaterial

Article 2(18) defines a nanomaterial as a natural, incidental, or manufactured material containing particles in an unbound state, as an aggregate, or as an agglomerate, where 50% or more of the particles — by number size distribution — have one or more external dimensions in the 1-100 nanometer range. An aggregate is a particle made of strongly bound or fused particles; an agglomerate is a collection of weakly bound particles or aggregates whose combined surface area resembles the sum of its parts. The definition then adds a deliberate exception: fullerenes, graphene flakes, and single-wall carbon nanotubes are deemed nanomaterials even where one or more external dimensions falls below 1 nm. A materials team screening only for particles inside the 1-100 nm band will miss that carve-out.

The three exposure tiers

Rule 19 itself is short: a device incorporating or consisting of nanomaterial is Class III if it presents a high or medium potential for internal exposure, Class IIb if that potential is low, and Class IIa if it is negligible. Nothing in the rule allows a return to Class I. The rule exists because the old directive's classification logic was built around invasiveness and duration of contact, categories that don't capture what a nanoscale particle can do once it crosses a barrier the bulk material never would have.

• Class III — high or medium potential for internal exposure. The device's nanomaterial can reach internal tissue, cells, or body fluids in meaningful quantity.
• Class IIb — low potential for internal exposure. Some pathway to internal exposure exists, but it is limited by design, location, or the nanomaterial's form.
• Class IIa — negligible potential for internal exposure. The floor outcome under Rule 19 — still up-classified from whatever the device would otherwise be, but not treated as a meaningful internal-exposure risk.

Internal exposure potential is not a self-assessment on vibes. It is a factor-by-factor argument — application site, contact type, contact time, and how the nanomaterial is actually held in the device — that has to survive a Notified Body's technical documentation review. Why the exposure assessment carries the weight

What actually decides the tier

MDCG 2021-24 sets out the factors a manufacturer has to work through to land on high, medium, low, or negligible: the device's application site, the type of contact (tissue, cells, or body fluids), the contact time, and — critically — how the nanomaterial is incorporated into the device. Internal exposure itself is scoped to specific routes: injured skin or mucous membrane, surgically invasive devices, and implantable devices. A device with no plausible route to any of those may still incorporate nanomaterial and still be up-classified under Rule 19, but its exposure tier will land lower.

1. Free nanomaterial. Not bound to the device matrix or surface — the most direct exposure pathway, and the one that draws the closest scrutiny.
2. Fixed nanomaterial, e.g. a coating. Here the interaction mechanism matters: MDCG 2021-24 distinguishes chemisorption (a stronger chemical bond to the surface) from physisorption (a weaker physical attachment more prone to release).
3. Embedded nanomaterial. Held within a matrix rather than on a surface, where the matrix's own degradability becomes the deciding question — a degrading matrix can release what a stable one never would.

Where this quietly bites

The devices that get caught out are rarely the ones a regulatory team was already watching for exotic materials. Nanostructured surface coatings on orthopedic and dental implants, particulate excipients carried into a device-side combination product, and antimicrobial wound dressings using nanosilver all incorporate nanomaterial by the Article 2(18) definition — and none of them read, on the surface, like a nanotechnology device. The failure mode is organizational as much as technical: a materials or supply-chain team changes a coating supplier or an excipient grade without flagging the change to regulatory, and the classification implication surfaces for the first time in a Notified Body's technical file review rather than in the design history file where it belonged.

None of this is exotic once it's on the table — it's a bounded materials review, a documented incorporation-type and exposure argument, and a classification file that can withstand a Notified Body reading it cold. What makes Rule 19 dangerous is upstream of the assessment itself: a device team that never asks the question because nothing about the product looks like "nanotechnology." The risk management file and the classification rationale need to ask it together, because a nanomaterial that changes your class also changes what your risk file has to justify. If your last classification review predates a materials or supplier change, that is where to start; our EU MDR classification strategy work exists for exactly this gap.