Sterile devices placed on the EU market must be produced by a validated sterilization process and the validation evidence must live inside the technical documentation. Annex I Section 11 of Regulation (EU) 2017/745 sets the obligation: devices labelled as sterile must be designed, manufactured, and packaged in appropriate, validated conditions and by appropriate, validated methods. Annex II Section 6 is where the evidence is filed — pre-clinical and clinical evidence, including process validation reports. The required structure is the same regardless of whether you sterilize by ethylene oxide, gamma or electron-beam irradiation, moist heat, or a filtration or aseptic process: installation qualification, operational qualification, performance qualification, a defined sterility assurance level, validated sterile barrier packaging, and a routine monitoring programme that keeps the validation live.

By Tibor Zechmeister and Felix Lenhard. Last updated 10 April 2026.

TL;DR

  • MDR Annex I Section 11 requires devices labelled as sterile to be manufactured and packaged under validated conditions by validated methods, with packaging that maintains sterility until the point of use.
  • The evidence that the sterilization process is validated is filed in Annex II Section 6 of the technical documentation, alongside the rest of the pre-clinical evidence.
  • A complete sterilization validation has three qualification stages — installation, operational, and performance qualification — and culminates in a documented sterility assurance level, typically 10 to the minus 6 for terminally sterilized devices.
  • The sterile barrier system is part of the validation, not a separate topic. Packaging integrity evidence lives in the same dossier as the process evidence.
  • Routine monitoring and periodic revalidation keep the validation status current. A validation report from three years ago with no routine data behind it is not a validated process.

Why this is the section Notified Bodies open first

When a sterile device arrives at audit, the Notified Body reviewer opens the sterilization section of the technical file early. The reason is simple. A device labelled sterile that was not actually produced by a validated process is a patient safety issue of a kind that travels fast through the vigilance system. Notified Bodies know which failure modes generate serious incidents and which do not, and sterility failures are near the top of the list. The auditor wants to see, before anything else, that the team understood the obligation and built the evidence to match it.

The file either answers the obligation or it does not. There is no middle ground on a sterility claim. If the validation is incomplete, if the sterile barrier has not been tested, if routine monitoring is missing, the finding is written on the first read. A startup that gets this section right rarely has a painful audit elsewhere, because the discipline required to document sterilization correctly tends to produce a file that is disciplined throughout.

Annex I Section 11 — what the regulation actually requires

Annex I of Regulation (EU) 2017/745 contains the General Safety and Performance Requirements. Section 11 of Annex I is the section on infection and microbial contamination. It is the direct legal source for every sterility obligation that applies to a medical device on the EU market.

Section 11 requires that devices and their manufacturing processes be designed to eliminate or reduce as far as possible the risk of infection to the patient, user, and, where applicable, other persons. Devices labelled as having a specific microbial state must be designed, manufactured, and packaged to ensure that they remain in that state when placed on the market and under the transport and storage conditions specified by the manufacturer. Devices delivered in a sterile state must be designed, manufactured, and packaged in accordance with appropriate procedures to ensure that they are sterile when placed on the market and remain sterile, under the storage and transport conditions specified, until the protective packaging is damaged or opened. Devices delivered in a sterile state must have been manufactured and sterilized by an appropriate, validated method.

Read that last sentence carefully. The regulation does not say the sterilization process should be validated. It says the device must have been manufactured and sterilized by a validated method. Validation is a precondition of placing the device on the market. No validation, no market.

Annex II Section 6 — where the evidence is filed

Annex II of Regulation (EU) 2017/745 specifies the structure of the technical documentation. Section 6 of Annex II is the pre-clinical and clinical evidence section. This is where sterilization validation documentation lives. Not in a separate binder, not as an appendix, not as a summary with the underlying reports archived somewhere else. In Section 6, controlled, version-stamped, cross-referenced from the GSPR checklist in Section 4 against Annex I Section 11.

The cross-reference discipline matters. An auditor sampling the GSPR checklist for Annex I Section 11 should land on the sterilization validation reports inside Annex II Section 6 without detour. If that trace breaks — if Section 4 of the file says the evidence lives somewhere and Section 6 does not contain it at that location — the finding is written before the content of the reports has even been read.

Sterilization method families you will choose between

The validation logic is consistent across methods, but the standards and the specific parameters differ by method. The main terminal sterilization families for medical devices are ethylene oxide, radiation (gamma, electron beam, and increasingly X-ray), moist heat, and dry heat. Devices that cannot survive a terminal process use an aseptic processing route or, for certain liquids and biologics, sterilizing filtration combined with aseptic filling.

Each family has its own harmonised standard with its own parameters and validation requirements. Ethylene oxide validation follows a different logic from gamma irradiation, which follows a different logic from moist heat. The Notified Body expects the dossier to use the correct standard for the chosen method and to follow the validation approach that standard specifies. Choosing the method is a design decision with multi-year consequences for the technical file, the supply chain, and the product shelf life. Changing the method later means rerunning the validation from scratch.

For startups, the practical point is that the sterilization method is usually dictated by the material compatibility of the device. A polymer that degrades under gamma is not a gamma candidate. A thermally sensitive electronic assembly is not a moist heat candidate. A device with long, narrow lumens may not be a good ethylene oxide candidate because of residual diffusion. The right move is to choose the method with the contract sterilizer and the packaging engineer in the room, not in isolation.

Validation approaches — overkill versus bioburden-based

Inside each method, there are different validation approaches defined by the relevant harmonised standards. The two concepts worth naming are the overkill approach and the bioburden-based approach.

The overkill approach applies a process lethality well in excess of what is required to inactivate a conservatively assumed initial microbial load. It is simple to validate and tolerant of manufacturing variability. It is also harder on the product, and it is not available for devices that cannot survive the conditions required.

The bioburden-based approach sets the process parameters based on measured bioburden in the device before sterilization and on the resistance characteristics of the organisms present. It is gentler on the product and more efficient in cycle time, but it requires a real bioburden monitoring programme in the manufacturing environment and disciplined routine control. It is not a shortcut. It is a commitment to a continuous data stream that the overkill approach does not demand.

The choice between approaches is documented in the validation master plan, which is itself a deliverable inside the technical file.

Process qualification stages — IQ, OQ, PQ

Sterilization validation, across every method, follows a three-stage qualification structure. These stages originate in the process validation tradition that ISO and CEN sterilization standards have built on for decades, and they are what a Notified Body reviewer expects to see in a sterilization dossier.

Installation Qualification (IQ) documents that the sterilization equipment is installed correctly and meets the manufacturer's specification, with calibration of all critical instruments, documented utility connections, and documented software versions where applicable. IQ is usually performed by the contract sterilizer for devices sterilized externally, and the IQ report is provided to the device manufacturer as part of the contract.

Operational Qualification (OQ) establishes that the equipment, operated within defined parameters, delivers a process within the required specifications. OQ defines the worst-case cycle conditions that will be used for subsequent performance qualification. OQ is typically equipment-specific and may be leveraged across multiple product families processed on the same equipment.

Performance Qualification (PQ) is where the product enters the picture. PQ demonstrates that the process, run under OQ conditions, achieves the required sterility assurance level for the specific device inside its specific packaging configuration in its specific loading pattern. PQ is product-specific. Half-cycle studies, biological indicator placement, dose distribution mapping — the specific techniques depend on the method, but the PQ report is the one that ties the validation to the device the manufacturer is placing on the market.

A validation dossier that documents IQ and OQ from the contract sterilizer but is missing a product-specific PQ for the device under audit is not a validated process. It is a partially validated facility.

The sterility assurance level

Terminally sterilized medical devices are typically validated to a sterility assurance level (SAL) of 10 to the minus 6. This means a theoretical probability of no more than one non-sterile unit in one million units processed. The SAL is a property of the validation, not a test result you measure on each batch. It is the level the validation demonstrates the process can achieve, and routine monitoring is what keeps the claim alive.

For devices that cannot withstand a 10 to the minus 6 terminal process — for example, some heat-sensitive or radiation-sensitive products — a less stringent SAL may be acceptable under specific conditions, typically with additional labelling or handling requirements. This is an exception territory and should be discussed with the Notified Body early, not justified in the final dossier.

Sterile barrier integrity — the packaging is part of the process

A perfectly validated sterilization cycle inside a packaging system that fails in transport produces an unsterile device on the hospital shelf. The sterile barrier system — the primary packaging that maintains sterility from the end of sterilization until the point of use — is therefore part of the validation, not a separate workstream.

The dossier must include evidence that the sterile barrier system maintains integrity through the validated sterilization process, through the declared shelf life, and through the transport and handling conditions the device will realistically see. Seal strength, burst strength, dye penetration, microbial barrier, ageing studies, and transport simulation all live in this part of the file. The shelf life claim on the label is backed by accelerated and, where required, real-time ageing data for the sterile barrier system. A shelf life claim with no ageing evidence is one of the fastest findings a reviewer can write.

Routine monitoring keeps the validation alive

Validation is not a one-time event that sits in the file forever. Sterilization processes are live manufacturing processes, and the validation status is maintained through routine monitoring and periodic revalidation. Ethylene oxide cycles are monitored with biological indicators and parametric release data. Radiation cycles are monitored with dosimetry. Bioburden is monitored on a defined schedule. Deviations are investigated, documented, and fed back into the risk management file.

Periodic revalidation — typically annually for radiation dose audits, at defined intervals for other methods — confirms that the validated parameters still deliver the claimed SAL under current manufacturing conditions. A three-year-old validation report with no routine data and no revalidation since is not a living validation. A Notified Body reviewer will treat it as an expired one.

How the documentation is structured inside the file

Inside Annex II Section 6, the sterilization validation documentation typically consists of: a validation master plan describing the overall approach, the IQ report, the OQ report, the PQ report, the bioburden monitoring programme, the sterile barrier system qualification including ageing data, the routine monitoring procedures and records, the revalidation schedule, and the contract sterilizer agreements where external sterilization is used. Each document is version-controlled under EN ISO 13485:2016+A11:2021. Each document is cross-referenced from the Annex I Section 11 line of the GSPR checklist. The dossier reads as a single evidence chain, not as a stack of unrelated reports.

Common gaps in startup sterilization dossiers

The recurring patterns we see when reviewing startup sterilization sections before a first audit:

  • Contract sterilizer reports treated as the whole validation. The IQ and OQ from the contract sterilizer are present, but there is no product-specific PQ for the device inside its actual packaging in its actual load configuration.
  • Sterile barrier system documented as an afterthought. The process is validated, the packaging is not, and the shelf life claim has no ageing evidence behind it.
  • Routine monitoring missing or fragmentary. The validation report exists. The last two years of dosimetry or biological indicator data do not.
  • No revalidation schedule. The file says the process is validated. It does not say when the validation will next be confirmed.
  • Bioburden data absent under a bioburden-based approach. The team chose the gentler validation approach without setting up the continuous monitoring programme that approach requires.
  • Sterility assurance level stated without derivation. The dossier claims a SAL of 10 to the minus 6 without showing the validation data that justifies the claim.
  • Transport simulation missing. The sterile barrier system has been tested statically but not under realistic distribution conditions.

Every one of these is a finding. None of them is rare.

The Subtract to Ship approach

Sterilization validation is an area where the pressure to add is intense. Extra biological indicators, extra dosimeters, extra packaging tests, extra ageing protocols, extra contract sterilizer paperwork. Most of it is not wrong — it is simply uncoordinated. A file that grows by accumulation ends up larger and less defensible than one built from a single validation master plan that maps every deliverable to a specific requirement in Annex I Section 11 and a specific reference in Annex II Section 6.

The Subtract to Ship framework for MDR applies the same test here it applies everywhere. Every document in the sterilization dossier must trace to a specific requirement and to a specific line in the GSPR checklist. Documents that cannot answer "what does this satisfy?" come out. What remains is a validation dossier a reviewer can follow in one pass and a team can keep current without heroics.

The startups that get this right are the ones who wrote the validation master plan before they ran the first cycle. The ones who pay twice are the ones who ran cycles first and tried to write the plan around the data afterwards.

Reality Check — Where do you stand?

  1. For your chosen sterilization method, can you point to the specific harmonised standard version your validation follows?
  2. Do you have a written validation master plan that defines the approach — overkill or bioburden-based — before the qualification runs started?
  3. Is there a product-specific Performance Qualification report for your device in its actual packaging, or is the validation leaning entirely on the contract sterilizer's facility-level reports?
  4. Does your GSPR checklist line for Annex I Section 11 cross-reference to specific documents inside Annex II Section 6, and do those references resolve today?
  5. Is the sterile barrier system qualification — seal strength, microbial barrier, ageing, transport simulation — filed alongside the process validation, or is it fragmented across the file?
  6. Does the shelf life claim on the label match the ageing evidence in the file, with both accelerated and real-time data for the claimed period?
  7. Is there a routine monitoring programme running now, with the last two years of data available on request?
  8. Is there a revalidation schedule with a named owner and the next revalidation date on the calendar?

Frequently Asked Questions

What MDR article requires sterilization validation? The obligation sits in Annex I Section 11 of Regulation (EU) 2017/745. Devices labelled as sterile must have been manufactured and sterilized by an appropriate, validated method, and the packaging must ensure the device remains sterile under the specified storage and transport conditions until the protective packaging is damaged or opened.

Where does sterilization validation evidence go in the technical file? In Annex II Section 6, the pre-clinical and clinical evidence section. It is cross-referenced from the Annex I Section 11 line of the GSPR checklist in Annex II Section 4. The validation reports themselves live as version-controlled documents under the QMS document control procedure.

Is a sterility assurance level of 10 to the minus 6 always required? For terminally sterilized medical devices, a SAL of 10 to the minus 6 is the standard expectation. For products that cannot survive that level of process lethality, a less stringent SAL may be acceptable in specific circumstances, usually with additional labelling or handling considerations. This is exception territory and should be discussed with the Notified Body early in the design phase.

Can a startup use a contract sterilizer's validation as its own? Partially. The contract sterilizer's Installation Qualification and Operational Qualification for the facility and equipment can be leveraged. The Performance Qualification for the specific device in its specific packaging configuration is the manufacturer's responsibility, and it must be performed with the actual product. A dossier that contains only facility-level validation from the contract sterilizer is not complete.

How often does sterilization validation need to be repeated? Revalidation frequency depends on the method and the harmonised standard used. Radiation dose audits are typically performed at defined quarterly or annual intervals depending on the approach. Ethylene oxide and moist heat processes have their own revalidation schedules. Any significant change to the device, the packaging, the load configuration, the supplier, or the equipment also triggers revalidation.

Does the sterile barrier system really need its own ageing studies? Yes. The shelf life claim on the label depends on evidence that the sterile barrier system maintains its integrity over the claimed period. Accelerated ageing is used to support initial claims, and real-time ageing data is expected to follow and confirm the accelerated results. A shelf life claim with no ageing evidence is a finding a Notified Body will write on the first pass.

What happens if routine monitoring data is missing from the file? The validation is treated as not current. A validation report from three years ago with no dosimetry, no biological indicator data, and no revalidation since is not a living validation. The reviewer will ask for the monitoring data on the spot, and if it is not available, the sterility claim on the label is not supportable.

Sources

  1. Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, Annex I Section 11 (infection and microbial contamination) and Annex II Section 6 (pre-clinical and clinical evidence). Official Journal L 117, 5.5.2017.
  2. EN ISO 13485:2016 + A11:2021 — Medical devices — Quality management systems — Requirements for regulatory purposes.

This post is part of the Technical Documentation cluster in the Subtract to Ship: MDR blog. Authored by Felix Lenhard and Tibor Zechmeister. Tibor has reviewed sterilization validation dossiers on both sides of the Notified Body table — as the lead auditor evaluating them and as a founder building his own. Where a single validated sterilization process meets a disciplined Annex II Section 6 structure, first audits on sterile devices go smoothly. Where either half is missing, they do not.