What Constitutes the Lifetime of a Medical Device in the EU?

Operating medical devices beyond the end of their lifetime can be dangerous and legally problematic. Therefore, manufacturers should precisely define the lifetime of every device. But, if you were put on the spot, would you be able to quickly explain how the lifetime is defined or what criteria are used to work it out?

If you couldn’t, you’re not alone: a lot of aspects regarding the lifetime of medical devices are unclear, especially at the European level.

Related terms such as “shelf life” and “operating time” often cause additional confusion – as does the fact that the lifetime does not necessarily have to be a “duration,” other factors can also play a role.

Avoid problems with regulatory authorities and patient-related costs, or even damages, and find out in this article:

• Why working out the lifetime is important
• How to define and establish the lifetime
• How lifetime differs from other concepts such as end of service or operating time
• Which (regulatory) requirements you must comply with as a manufacturer and operator

1. Why the lifetime and its definition are important

The lifetime of a medical device is the period during which the device is safe to use and performs as intended. Naturally, a device's properties change over time, for example, due to aging or wear. Major problems can arise if these changes mean that safety and/or performance are no longer guaranteed. For example: 

• A dialysis machine's fluid balancing becomes inaccurate over time as the sealing materials age, resulting in fluid losses.
• Plastics become brittle and lose their fracture strength.
• Electronic components can malfunction after a lot of switching operations.

The lifetime of medical devices is important for both manufacturers and operators.

a) Importance for manufacturers

Manufacturers have to know the lifetime when formulating the intended purpose.

• They need the lifetime for the risk management process (Annex I MDR, ISO 14971), where they need to analyze whether the device is safe and performs as intended over its entire lifetime.

• The lifetime correlates with the total number of uses. Both affect the risk acceptance matrix. The class of the least probable harm (usually the bottom row) should have probabilities that mean they shouldn’t be expected to occur during all uses over the complete lifetime of the device.
• The intended lifetime conditions the selection of components, materials and technologies even during development.
• The lifetime and necessary service intervals are closely linked to one another. The manufacturer documents these service intervals in the accompanying information. 

b) Importance for operators

Operators also have to act at the end of the device's specified lifetime (e.g., by taking the device out of service). Operators often have a hard time with this because the device seems to be working flawlessly. They cannot see how the safety or performance is compromised. This is probably also the reason many operators struggle with decommissioning. 

It is even more difficult if the manufacturer has not specified the lifetime in the accompanying documentation. In this case, how is the operator supposed to know when safety and performance are no longer guaranteed?

Even if operators think their device is still working perfectly: the lifetime specified by the manufacturer takes precedence.

c) Importance for customers

The lifetime of a device is also an important factor for customers. Indeed, in a lot of cases, it is one of the factors taken into consideration when deciding whether to purchase a device. If a purchaser at a hospital purchaser has a choice between two otherwise similar devices, they will choose the one with the longer lifetime.

2. The complicated definition of lifetime

As important as it is for manufacturers, operators, customers and authorities to know what the lifetime of a medical device is, it is just as difficult to precisely define (and thus establish) what constitutes the “lifetime.” Legal regulations mostly remain vague, and a lot of terms are used synonymously with “lifetime.”

a) MDR

EU Regulation 2017/745 (MDR) does not explicitly define lifetime. However, in Annex I, it does summarize the general safety and performance requirements and indirectly and incidentally define the term “lifetime”:

“As indicated by the manufacturer” refers to “the lifetime of the device” and this enables us to understand the requirement for the manufacturer to specify the lifetime of the device.

b) IMDRF

While the MDR does not explicitly define the term lifetime, there is a more detailed explanation of the concept in the “Essential Principles of Safety and Performance ” document from the International Medical Device Regulators Forum (IMDRF). According to this explanation, the “expected lifetime” is: 

Section 2.3 of the old IMDRF document “GHTF SG2 N21 R8 May 1999” made the “service life”, as it called it, dependent on the intended duration of use:

c) Active medical devices: IEC 60601-1

For active medical devices, IEC 60601-1 defines the “expected service life” as the:

d) How you should deal with the different definitions

As evidence of compliance with the MDR's general safety and performance requirements as defined in MDR Annex I, paragraph 6 has to be provided in the device's technical documentation, we recommend documenting your definition of the term “lifetime” there as well.

One manufacturer might think of the lifetime as the number of uses (“300 uses”), another as a time period (“up to 5 years after commissioning”) or up to a specific date (“until 12/2025”). Combinations are also possible: 5 years or 1,000 treatments, whichever comes first.

3. What the lifetime is NOT (and what you nevertheless need to pay attention to)

It is also important to know which terms, although often used in connection with lifetime, should not be confused with it.

a) Best before date

Most of us are familiar with the idea of a "best before date” from the food industry. But this term is not (or is no longer) used in the medical device sector. In contrast to foodstuffs, where the best before date is more of a guide, the shelf life and lifetime of a medical device are non-negotiable. 

This means that after the end of the specified period or when a specified expiry date has been reached (see ISO 15223: “use-by-date”), safety and/or performance are no longer guaranteed and, therefore, the device may no longer be used.

b) Stability, including shelf life

In ZLG Document 3.3 A 5 (Shelf life – Evaluation of the defined shelf life) shelf life is defined as:

Examples 

• An implant may have a lifetime of 15 years (including time in the human body), but a shelf life of only 2 years in terms of stability in its original packaging
• A substance-based medical device may be stored in its original packaging for three years (shelf life) but have to be used within 12 months of being opened (in-use stability)

Stability of IVD reagents 

The stability of reagent devices, including calibration and control materials for in vitro diagnostic medical devices, is particularly important. They are generally very susceptible to environmental influences such as light, temperature or vibration.

The stability testing must be part of the technical documentation according to Annex II, 6.3 of the IVDR. 

• Relevant standard: ISO 23640 “Evaluation of stability of in vitro diagnostic reagents”
• Relevant guidance document: CLSI guidance document EP25-A

ISO 23640 is expected to be harmonized with the IVDR. The standard is supplemented by CLSI guidance document EP25-A. Both require real-time testing under realistic conditions. However, they also allow for “accelerated stability testing” to enable an initial estimate of the stability and shelf life of a device. 

c) End of service

Although the terms are similar, the end of service is not aligned with the lifetime.

This period is determined by the manufacturer or the service organization and is not defined from a regulatory perspective unless the manufacturer cancels services before the end of the device's lifetime. In this case, the medical device can no longer be operated safely and perform as intended due to a lack of maintenance and repair and may have to be taken out of operation or withdrawn from the market by the operator.

d) Operating period

The operating period is the name given to the period of active operation itself, i.e., the period when the device is actually used. This is not the same as the lifetime, as the lifetime does not necessarily depend on actual use.

Example:

A dialysis machine has a lifetime of ten years. Even if the device is only used for half an hour during these ten years, the lifetime ends at the end of these ten years.

The operating period, therefore, depends less on the device than it does on the area of use and its users. It often correlates to a certain extent with the lifetime, but not always because downtimes, storage and transport also cause aging or other effects.

Because the age-related and the use-related lifetime can be different, as mentioned above in some cases both are specified: 5 years or 1000 treatments, whichever comes first.

e) End of life

The IMDRF's document on cybersecurity defines the end of life as:

So, it is the time at which a manufacturer ceases to market a device. Therefore, “end of life” refers to a product line not, as in the case of the lifetime, an individual device!

However, the US FDA uses the term “end of life” (EOL) differently. In 21 CFR 803.3, subparagraph (f) states:

f) How you should deal with the different terms

Our tips:

• Use the terms used in the regulatory requirements
• List the terms in your documentation. Explicitly provide definitions again in a glossary
• Provide a reference to the origin of the definition in your glossary or state that it is your own definition
• Be aware that there may be several aging and lifetime-related aspects that are all relevant for your device. Corresponding proof is required for each individual aspect

4. Determining the lifetime

a) Determining the lifetime of physical devices

General criteria for determining the lifetime

Unfortunately, legal regulations such as the MDR and IVDR do not provide any direct guidance on how the lifetime of a medical device should be determined. 

Therefore, when determining the lifetime of a physical device, manufacturers should use relevant standards as a guide. However, there is no standard that covers all cases. 

The now-withdrawn TR/ISO 14969:2004, which still referred to the application of ISO 13495:2003, had a very helpful summary of how to define the lifetime of medical devices in section 7.1.3:

The standard also described other factors that should be considered when defining the lifetime: 

In addition, commercial software products that can model lifetime are also available.

Special standards for individual topics

There are up-to-date standards for specific topics. We recommend performing a detailed search of standards databases, e.g., iso.org or its full text search, to identify these standards. 

Examples of specific standards are:

• ISO 11346:2014 Rubber, vulcanized or thermoplastic — Estimation of lifetime and maximum temperature of use

• ISO 17526:2003 Optics and optical instruments — Lasers and laser-related equipment — Lifetime of lasers
• ISO/IEC 16963:2017 Information technology — Digitally recorded media for information interchange and storage — Test method for the estimation of lifetime of optical disks for long-term data storage

There are also standards describing aging tests for various materials. Search for “accelerated aging test” using the standards database’s full-text search.

Examples of standards on aging tests:

• ISO 188:2011 Rubber, vulcanized or thermoplastic — Accelerated aging and heat resistance tests
• ISO 23640 Evaluation of stability of in vitro diagnostic reagents

Maintenance

There may be opportunities to check aged components through regular tests prior to use or through self-testing by the device and thus to detect and counteract unsafe conditions or reduced device performance in good time.

Using information on aging and from risk management, maintenance strategies that will extend the specified lifetime can be developed.

b) Determining the lifetime of software devices

Software itself does not age. But the software's environment does change: new operating systems, new technologies, new communication protocols, etc.

A pure software product will therefore rarely have a lifetime longer than a few years without modifications. Five years would be a good rule of thumb, although it should be calculated based on risk.

C) Tips for determining product lifetime

There are hardly any guidelines in the west on how to determine product lifetime but a guidance document from China has compiled some information. We have taken the following tips from it to help you to determine your product lifetime: 

Step 1: establishing the desired product lifetime in the requirements specification

Prior to or at the beginning of the development stage, manufacturers should determine the product lifetime.

Indications of how long this period should be can be derived, for example, from the following considerations:

• commercial considerations: comparison with competitors’ products; customer requirements; profitability, etc.
• technical considerations: feasibility and time until components wear out
• empirical data: empirical values from current market surveillance and test results

As one of the stakeholder requirements, the lifetime is a specification and not a result of development. Rather, the product management department puts forward a proposal regarding the lifetime along with the stakeholder requirements. Via risk management, technical data sheets or empirical tests, the development team assesses the lifetime proposal and confirms it, raises any issues or offers solutions (e.g. preventive maintenance, more expensive components, etc.).

Step 2: creating a system and component design

The product is designed with the previously established requirement specifications in mind. Here, among other things, suitable technology and product components need to be identified that meet the requirements necessary to achieve the lifetime and fulfill all other product requirements.

The design should demonstrably include the calculations and derivations for selecting or designing certain components. On the basis of the calculations, an auditor would decide whether or not certain tests need to be carried out. Good documentation also helps to limit the scope of the tests.

The following aspects should be included in the design documents:

• derivation of the lifetime
• list of critical components
• reliability calculations or results of simulations
• calculations of safety factors according to Table 21 of IEC 60601-1
• environmental conditions that affect the lifetime
• results of the risk assessment
• if necessary, test requirements to prove the lifetime
• comparison with reference designs
• standards that can be observed

Step 3: analyzing the risks associated with the product lifetime and implementing measures

In risk management, manufacturers must identify the risks to the product and components that are linked to safety and performance beyond the lifetime of the components.

Here, the following points are particularly to be considered:

• the product itself
  e.g. friction between two movable parts, actions of force
• the usage criteria
  e.g. frequency and intensity of use, maintenance and repairs
• the environment
  e.g. transport, storage, operating environment (temperature/humidity, UV light)
• cleaning and disinfection
• combinations of different influencing factor

Cleaning and disinfection

Cleaning and disinfection are factors that are frequently overlooked when determining the product lifetime. Use of the product may involve cleaning and disinfection. The cumulative effect of heating or drying procedures and chemical residues from such procedures can affect the product’s performance and should be assessed in terms of the effects on the product lifetime.

Different sterilization methods (steam sterilization, ethylene oxide sterilization, radio sterilization, etc.) and packaging methods can also have different effects on the product lifetime. Therefore, for example, the effects of the sterilization procedure (methods and parameters) on the product lifetime should be considered.

Ways of extending the lifetime

For each component, the materials and operating principles must support the required lifetime. If this is not possible, the following options can guarantee the component lifetime and minimize risks:

• option 1: maintenance
  Manufacturers can allow for preventive replacement of components. In this case, the corresponding maintenance intervals must be included in the instructions for use (IFU).
• option 2: test function
  Alternatively (or additionally), manufacturers can implement a test function on the product that reliably detects component failure (or other safety and performance limitations). This is always an option if the test function makes the product safer. However, it also means that this option is not effective in the case of life-sustaining/life-support systems. If their performance is limited or they malfunction and the test function is triggered, it is already too late.
• option 3: reducing the product lifetime

If neither of the other options is feasible, all that remains is for manufacturers to reduce the product lifetime and adapt it to the requirement specifications.

5. Regulatory requirements for the lifetime of medical devices

There are no legal regulations that exclusively and conclusively deal with the lifetime of medical devices. Instead, it is always individual aspects that are regulated. These include in particular:

• Full refurbishment and effect on lifetime
• Labeling of lifetime
• Quality management system and lifetime
• Communication of the lifetime
• Active medical devices and lifetime

a) Full refurbishment lets the lifetime start afresh

This means medical devices can be restored to a condition that is either the same as the original condition of a new device or to a condition for which a new lifetime has to be specified. 

The manufacturer must decide and document the effect of the refurbishment individually in each case. Cleaning, disinfection and sterilization aspects may also have to be considered. ISO 17664 can provide some guidance in this regard.

Full refurbishment may also mean a medical device no longer has to be disposed of. Whether or not this makes financial sense is up to the operator or the manufacturer to decide.
Operators generally rely on the manufacturer’s expertise when deciding on the scope, methods and details of a full refurbishment. However, there is no legal obligation for them to provide advice, which can be a challenge for operators.

b) Lifetime must be specified in the accompanying documents

For implantable devices, the MDR states:

This is the only place in European regulations that explicitly requires the lifetime of a device to be stated in the information accompanying a device. However, other regulations in the MDR refer to information regarding lifetime in the accompanying documents.

Annex I MDR, section 23.4: information in the instructions for use

This means that, in addition to the replacement of wear parts and regular inspection of medical devices during maintenance, calibrations, which the manufacturer must define, also play a role.

Annex VI MDR, Part B: the UDI system

This section is about the labeling of devices but is more concerned with the interpretation of the labeling. The manufacturer must be able to demonstrate that the information plate with UDI is permanently affixed to the device, is always legible and withstands all external influences it is exposed to when the device is used as intended. The selection of materials for the label plays just as important a role as the printing ink and technique, light and temperature influences, fluids and humidity, vibrations and mechanical loads or the choice of appropriate disinfectants.

With regard to the permanent legibility of information plates, it must be ensured that the replacement of the device components the information plate is affixed to is technically and procedurally possible. 

Have technicians in the field been informed of the situation? Can new information plates with a device's original data be generated on request? Can they be affixed in the field? How will they be verified in that case? It is these seemingly trivial special cases that rip holes in otherwise flawless regulatory compliance.

c) Lifetime conditions the PMS and PMCF

Article 83 MDR: post-market surveillance

The MDR establishes requirements for post-market surveillance (PMS):

In this case, lifetime seems to refer not only to the individual device but to all devices on the market. This point is particularly important if the manufacturer has already discontinued the device (see “end of life”) but must continue to carry out post-market surveillance, until the last device on the market has reached the end of its defined lifetime to be exact. If it can be demonstrated that there are no devices still on the market, the post-market surveillance can also be ended.

Article 86 MDR: periodic safety update report

This also applies to all devices on the market and describes the obligation to collect and evaluate specific information and to summarize it in a regular report.

Annex XIV, MDR: post-market clinical follow-up

The MDR also requires a post-market clinical follow-up (PMCF). This also depends on the lifetime:

As the lifetime is, by definition, the period of time during which a medical device can be operated safely and performs as intended, it is necessary to collect sufficient data throughout this period. This enables the manufacturer to demonstrate that all its assumptions and specifications were justified.

d) Communication of the lifetime

Users and operators must know from what point they may no longer operate the device because its safety and performance are no longer guaranteed. This alone makes it logically necessary for the manufacturer to communicate the lifetime.

But what are the mandatory legal requirements for this?

MDR

In fact, Article 18 (1)(c) of the MDR only requires a mandatory indication of lifetime in the case of implantable devices (see above).

IMDRF

Section 2.3 of the IMDRF document "GHTF SG2 N21 R8 May 1999" states:

“Where appropriate” can only be understood to mean that there may also be devices with a practically unlimited lifetime for which such information would make no sense. The same applies to sterile disposable products, for which the shelf life is more relevant than the lifetime since they are usually used immediately and then disposed of.

IEC 60601-1-11:2015

IEC 60601-1-11:2015 explicitly requires the expected service life to be specified in the documents accompanying medical electrical equipment for use in a domestic environment.

In general, it would be correct to say that, in most cases, a medical device's lifetime absolutely must be specified in the accompanying documents and, in some cases, it is even required by law or regulation. In any case, it must be included in the technical documentation.

e) Active medical devices according to IEC 60601

In IEC 60601-1, the lifetime is particularly relevant in connection with the term “single fault safe”.

Single fault safe is an important concept in IEC 60601-1. The standard defines it as:

Single fault safe

Manufacturers can only comply with this requirement if they have defined the service life and can demonstrate it with empirical tests and risk management in an appropriate and transparent way.

Edition 3.1 of IEC 60601-1 makes the following plea to manufacturers:

f) Requirements in China

Manufacturers are currently getting some nasty surprises when trying to get their devices authorized in China. Lifetime plays such an important role there that there is even separate NMPA guidance for it. Its title, translated literally, is Guideline for the Technical Review of the Lifetime of Active Medical Devices (No. 23 of 2019)”.

When registering medical devices, applicants must provide evidence to support the lifetime of their device as specified in the accompanying documents.

The regulations describe in detail how the lifetime should be defined and how it is taken into account in risk management.

The requirements are basically the same as the regulatory requirements in Europe, but here, in contrast to China, it is much less likely to become a point of attack during the authorization process. 

6. Conclusion

A medical device's safety and performance are only guaranteed during its specified lifetime. All aspects that affect risk must be taken into account when defining the lifetime. Unfortunately, a lot of manufacturers and operators take some liberties with the concept, in part because a longer lifetime can be economically advantageous and the definitions are fuzzy. 

Terms such as lifetime, shelf life, end of service, operating time and end of life are not synonyms and they have to be understood precisely if you want to comply with the regulatory requirements.

In Europe, lifetime plays a major role in relevant documents, such as the risk management file and test documentation. IEC 60601-1 provides valuable information on how to implement this in practice.

Markets such as China also require the lifetime to be specified in the accompanying documentation, and also require information on how it was calculated as well as reliable proof in the technical documentation.

Manufacturers in Europe would therefore do well to specify the desired lifetime as a design input in the future. They should consider it an “essential design output” (as defined by 21 CFR 820.30d) to ensure closer attention is paid to the actual lifetime of their devices in the lifecycle process and also provide sufficient evidence to support their claims.