IVDR Annex VIII's Rule 2 covers devices for blood grouping and tissue typing — and most classification guides summarize it in one line: blood grouping is Class D. That line is half right and the half that is missing is where new assay teams get surprised. Rule 2 names five specific marker systems that land in Class D. Everything else it covers — the rest of blood grouping, foeto-maternal incompatibility testing, HLA typing — defaults to Class C. The difference is not a technicality; it is the difference between a device that needs EU Reference Laboratory batch verification before it reaches the market and one that does not.

What Rule 2 actually covers

Rule 2 sits in Annex VIII's device-purpose rules, immediately after Rule 1's transmissible-agent default. Where Rule 1 catches devices that screen for infectious-disease risk in donated blood and tissue, Rule 2 catches devices that determine immunological compatibility — whether a given unit of blood, a component, a cell population, a tissue, or an organ is safe to give to a particular recipient. That is a narrower clinical question than Rule 1's, and the IVDR answers it with a narrower, more specific classification test: not "does this device touch blood safety broadly," but "does this device test for one of five named systems."

5 systems
ABO, Rhesus, Kell, Kidd, Duffy — the closed list that decides Class D under Rule 2.
Class D
Applies to determination of any of the five systems, in donor or recipient, including reverse ABO typing.
Class C
The default for every other blood grouping, foeto-maternal incompatibility, or HLA typing device Rule 2 covers.

The five-system list is closed, not illustrative

The most common misreading of Rule 2 treats its five named systems as examples of "major" blood groups, with the implication that a sufficiently important new marker would also qualify for Class D by analogy. It would not. The European Medicines Agency's Medical Device Coordination Group Expert Panel, asked to advise on where indirect antiglobulin tests fall, worked from the same premise the rule's text supports: Class D turns on whether a device determines ABO, Rhesus, Kell, Kidd, or Duffy — full stop. The Rhesus branch itself reaches further than many teams assume: it covers not just RhD but weak D, C, E, c, and e, and it covers the determination of partial D, because patients who are D-antigen positive by partial-D variants carry a real risk of anti-D alloimmunization that the device has to resolve correctly.

  • Reverse ABO typing counts. Detecting anti-A and anti-B antibodies to confirm a patient's ABO group from the antibody side, not just the antigen side, falls inside the Class D scope — it is still an ABO determination.
  • Antibody screening for the named systems is the harder edge case. Whether an indirect antiglobulin test is "determining" Kell, Kidd, or Duffy markers specifically, versus screening broadly for irregular antibodies, is exactly the question MDCG guidance has had to clarify case by case — not a line a sponsor should resolve by inference from the rule's text alone.
  • A new or emerging marker system is Class C until the rule says otherwise. Annex VIII is a closed list by design; a device testing for a blood group system outside the five named ones does not get Class D by argument from clinical importance.
  • If another rule also applies, the higher classification governs. Annex VIII's general principle — when several rules could apply to one device, the higher class wins — still operates underneath Rule 2's own split.
Rule 2 is not asking how clinically important your assay is. It is asking one narrow question: does it determine one of five named systems. Everything else about the device's risk profile is beside the point for this particular classification step. Why the closed list matters before you draft the intended purpose

What the classification decision actually changes

A Class D outcome under Rule 2 is not a paperwork distinction. It routes the device into the same Class D conformity-assessment track as a Rule 1 transmissible-agent test: notified body involvement, a scientific validity report built to the higher evidentiary standard, and — once a category's EU Reference Laboratory is operational — batch verification before the device reaches the market. A Class C outcome skips the EURL step entirely and sits on a lighter notified-body pathway, with its own conformity-assessment transition deadlines to track instead. Getting the Rule 2 call wrong in either direction has a real cost: understating it exposes the device to a non-conformity finding late in review; overstating it buys a verification burden the device's actual classification never required.

A Rule 2 classification check worth running before you draft the intended purpose
  1. Name the exact marker, not the clinical category. Write down whether the device determines ABO, Rhesus (including weak D and partial D variants), Kell, Kidd, or Duffy specifically — not whether it is "a blood grouping test."
  2. Check both the antigen and antibody direction. Reverse typing and antibody detection for the named systems sit inside Class D scope the same as forward typing does.
  3. Confirm donor-side and recipient-side use both. Rule 2 does not distinguish by which side of the transfusion or transplant the test is run on.
  4. Treat antibody-screening edge cases as a question for current MDCG guidance, not an inference. Indirect antiglobulin testing and similar screens have needed explicit Expert Panel clarification before — don't resolve a borderline case by analogy alone.

None of this changes what Rule 1 already does for infectious-disease screening devices, and it does not reach in vitro diagnostics outside the blood-grouping and tissue-typing purpose altogether. What it does is give assay teams a concrete first test to run before the classification memo gets written: does the device's intended purpose name one of five systems, or doesn't it. That single check decides whether the EURL timeline belongs in the project plan from day one.

Frequently asked questions

Which blood group systems trigger Class D under IVDR Rule 2?

Rule 2 names five systems: ABO, Rhesus (including RhD, weak D, C, E, c, and e), Kell, Kidd, and Duffy. A device intended to determine any of these — in a donor or a recipient, including reverse ABO typing by detecting anti-A and anti-B antibodies — is Class D. The list is closed, not illustrative: it is the specific set of systems ISO Blood Transfusion requires for safe transfusion decisions, not a general description of "major" blood groups.

Is HLA typing automatically Class D?

No. HLA typing falls under Rule 2's Class C branch alongside blood grouping or foeto-maternal incompatibility testing that does not involve the five named systems. A device can carry real clinical consequence — transplant compatibility, maternal-fetal risk — and still land in Class C if it does not test for ABO, Rhesus, Kell, Kidd, or Duffy markers specifically.

Does Rule 2 apply to both donor and recipient testing?

Yes. Rule 2's scope covers devices intended to ensure the immunological compatibility of blood, blood components, cells, tissue, or organs for transfusion or transplantation, and it applies the same classification logic on both sides of that exchange — testing the donor unit and testing the recipient both fall under the rule.

Sources & further reading

  1. Regulation (EU) 2017/746 of the European Parliament and of the Council on in vitro diagnostic medical devices, Annex VIII, Chapter III, Rule 2. eur-lex.europa.eu
  2. European Medicines Agency / Medical Device Coordination Group Expert Panel. Advice on indirect antiglobulin tests (request to the Medical Device Coordination Group), referencing MDCG 2020-16 Rule 2 classification criteria. ema.europa.eu

This article is provided for general informational purposes and reflects the regulatory landscape as of November 2026. It is not legal or regulatory advice. Confirm current IVDR Annex VIII classification guidance and MDCG interpretations with the European Commission, your notified body, or qualified counsel before relying on a specific classification determination.