Knowing and defining the expected lifetime of medical devices is essential for conformity assessment and for the safe operation of medical devices. The Medical Device Regulation (MDR) explicitly requires the specification of the lifetime in different places, e.g. for implants (Article 18 c)), in the "post-market surveillance" (Article 83) or in the "periodic safety update report" (Article 86). Essentially, the medical device must not pose any additional risk throughout its lifetime. In our consulting work, we repeatedly find that defining the service life can be a difficult undertaking. For this reason, we provide you here with practical advice on the elements to be taken into account in the definition.
Definition of "expected life"
How is the expected lifetime of medical devices defined?
Unfortunately, there is no explicit definition of the term in the Medical Device Regulation (EU) 2017/745 (MDR). However, Annex I, Chapter I, No. 6 states that "The characteristics and performance of a device shall not be adversely affected to such a degree that the health or safety of the patient or the user and, where applicable, of other persons are compromised during the lifetime of the device, as indicated by the manufacturer, when the device is subjected to the stresses which can occur during normal conditions of use and has been properly maintained in accordance with the manufacturer's instructions" (EU) 2017/745 MDR Annex I, Chapter I, No. 6.
Or more simply, "The expected lifetime is the period of time during which the medical device performs its function under normal conditions according to its risk and benefit profile, without posing any additional risk to the patient."
In the context of this paper, "expected lifetime" starts when the product is put into service by the user. From that point on, this time expires, mutatis mutandis.
Note: In other contexts, the term is sometimes used somewhat differently, e.g. as "expected service life" in DIN EN 60601-1.
In addition, the shelf life is also important. For example, in the case of sterile products, it describes the point in time up to which the product can be used safely (property: sterility). Here, the expiration of time starts with the completion of the packaging and sterilization process.
Shelf life
The shelf life defines the period of time until which the product remains suitable for its intended use (under appropriate storage and transport conditions).
Until this defined time, the product can be expected to retain its defined properties (for example: sterility, safety, performance).
This time period may be different from the "expected life". Example: for a device with internal power supply, the "shelf life" could be shorter than the "expected life" because the internal power supply (the "batteries") could discharge during storage.
This length of time can be determined by objective evidence, for example by validating the product design, the sterile process used, and the transport and storage conditions. Material properties have an influence, so that material degradation over time, for example, would have to be taken into account. Shelf life is also influenced by packaging, storage conditions (temperature fluctuations, humidity, radiation exposure) and other factors.
For certain product types, such as sterile products, the expiration date is a fixed quantity. Once the specified period has been exceeded, it is no longer certain that the product is sterile, even if the packaging is undamaged. This is because it is not possible to actually test whether the product is still sterile without compromising this property.
In the case of other products (example: batteries), on the other hand, suitable tests could determine whether they still have sufficient capacity and can be used (with restrictions, if necessary) even if the stated expiry date has passed (soft limit).
Software can also have an "expiry date"
Even if it sounds surprising: even "pure" software products (Software as Medical Device - SaMD) can have a limited shelf life and thus an expiration date. This is just as true for SiMD ("Software in Medical Device", i.e. "embedded software").
On the one hand - we will discuss this further below - external factors can lead to the software no longer being able to be installed and executed. In this case, the software on the shelf would have reached or exceeded its expiration date.
But keep in mind: any cryptographic certificates that have a validity period and thus an expiration date (!) also lead to a limited shelf life - also of SaMD or SiMD.
Lifetime (operating time until failure)
By determining the MTTF (Mean-Time-To-Failure) in the context of a reliability test, the average period of time can be found up to which the product (statistically speaking) shows a failure. It is therefore also referred to as "mean time to failure".
In addition, the MTTFD is known, the mean time to dangerous failure.
To the extent that device failure poses a risk to the patient, MTTF and MTTFD are to be considered equivalent, in terms of our definition of lifetime as: "period of time during which the medical device performs its function under normal conditions in accordance with its risk and benefit profile without posing an additional risk to the patient" (see above).
Note: MTTF/MTTFD are statistical quantities, they say nothing about an individual product.
They are determined via trials or empirical values. This involves simulating the use of the medical device in its entirety or, alternatively, defined, relevant functions, if necessary in cycles or periods, based on the expected (realistic / practice-relevant) use scenarios.
These tests, also known as durability tests, provide information on "how long it takes for a defect to occur in the product".
Inspection, maintenance and repair extend the service life
For a sterile product, the "sterile" property ends at the latest when the packaging is opened; for a product that cannot be repaired, the service life ends when the MTTF/ MTTFD is reached. An example of such a product is the incandescent lamp.
In the case of repairable products, on the other hand, preventive measures can be taken to extend the expected service life. During regular inspection, it can be determined whether the characteristics of the product have changed. Following this, maintenance (e.g.: replace internal power sources, lubricate) or repair could be performed. Preventive maintenance, on the other hand, performs such activities preventively, even without the need for the product's characteristics to have changed. Example: preventive replacement of wear parts before wear can affect the properties of the product.
The following procedure helps: the intervals for preventive maintenance / repair are selected in such a way that, with an appropriate safety factor, they are reliably within the time span before the MTTF/MTTFD is reached.
Through the inspection / preventive maintenance / repair, the time span now neither restarts, because the product again has all defined properties comparable to a new product.
Note: one speaks with repairable products also of the MTBF (Mean-Time-Between-Failures).
However, the responsibility for implementation does not lie solely with the manufacturers!
The operator is responsible
If the service life is defined by the manufacturer to be longer than the time to reach failure (i.e. MTTF/MTTFD), the necessary measures to ensure the Essential Safety and Performance Requirements (inspection (also: safety or metrological checks) / preventive maintenance / repair) must be communicated to the operator
Because these preventive and maintenance measures must be carried out by the operator! If the operator does not comply with these requirements, risks may arise.
Special case: Software
The considerations made for MTTF/MTTFD do not apply to software, because software does not wear out in the same way as hardware.
Nevertheless, it is important to define the "expected life" for software and especially for SaMD. But how, if the software does not age or wear out through use, i.e., the execution of code? When defining the expected service life, it must be taken into account for software that it ages due to external factors.
One such factor could be the operating system used, for example. At a certain point, the updated operating system (think of necessary patches with regard to security incidents!) no longer supports the medical software product - the end of life has been reached!
The same applies to newer hardware, which no longer supports the software product and thus means the end of life of the software product.
To avoid surprises, you as the manufacturer should define the expected service life of your software product independently of such external influences, for example by specifying that the service life of the software product ends when support is discontinued. The end of support could now be defined as a fixed period of time (until a certain date: day/month/year) or as a period of time after the appearance of a new major release.
Because (see above): within the expected lifetime the basic security and performance requirements must be met - the product must not develop any additional risk within the expected lifetime!
Contact us. We will be happy to work with you to develop the necessary procedures, provide templates, accompany you during implementation and support you in all regulatory matters.
