Guidance on how to interpret "significant change" of a medical device: General principles

On this page

• Overview of the general principles
• Where to start
• Using risk assessments
• Using the four-corner principle
• Design verification and validation

Overview of the general principles

A significant change under section 34 of the Medical Devices Regulations is a change that could reasonably be expected to affect the safety or effectiveness of a licensed Class III or IV medical device.

All devices listed on a Class III or IV licence will be reviewed in accordance with section 10 to 20 of the regulations.

A proposed modification to a Class II licence may require a Class III or Class IV amendment before the proposed modified device is sold if the:

• device on the Class II licence is intended to be used with a separately licensed Class III or IV medical device and
• proposed modification may impact the safety or effectiveness of the devices when all components are used together as intended

Any change made to a device, including changes to labelling and packaging, is likely significant if:

• the safety or effectiveness of the device could be affected
• the change was made in response to new information that alters your understanding about the safety or effectiveness of a medical device
• a new, previously unidentified hazard or hazardous situation was identified
• the likelihood of a known hazard occurring has changed
• the severity of an adverse consequence has changed
• a risk control measure has been added, removed or relaxed

If you are proposing multiple changes, you should consider each change individually and evaluate the combined impact the changes have on the device.

A determination that a change is significant may occur when:

• one of the proposed modifications is determined to be a significant change or
• the collective impact of any combination of proposed changes is deemed to be significant

In general, a determination that a change is non-significant may occur when:

• a risk assessment of the proposed changes concludes that
  • any new or modified risks are negligible and
  • the probability of introducing unintended consequences is negligible

In all cases, if any change is determined to be significant under any section of this guidance document, the final determination should be that an amendment application is required.

For significant changes, you must submit an application to amend your medical device licence as per section 34(a) of the regulations.

Where to start

The first step is to determine which section best describes your proposed change.

The type of change that best describes your proposed change should serve as your starting point in determining significance. This is only a starting point as there may be multiple sections that apply to your proposed change.

To help determine whether a change is significant:

• consider this general principles section as well as the applicable descriptions of the types of changes
• consider the impact the change may have on the patient or intended users
• consider if any specifications have changed
• determine whether the change, prior to testing, could have reasonably been expected to affect the safety, quality or effectiveness of the device, in either a positive or negative way
• consider the results of a risk assessment specific to the change
• consider whether data supporting the modified device can be interpolated from existing datasets
• consider the need for and results of verification and validation testing

Use this guidance document to assess each proposed change on its own, as well as the combined impact of all the changes, when you are:

• considering several simultaneous changes in the evolution of a licensed device
• modifying 1 or more component parts of a medical device licence as a system, test kit, group, family or group family

It may be useful to compare the proposed changes side by side with the current licensed device.

Using risk assessments

A risk assessment should be incorporated throughout the process of assessing the significance of a change and again as a final step.

A risk assessment is especially helpful in determining whether a change is significant in situations where the descriptions provided in this guidance may not:

• explicitly address the specific change or
• initially lead to a determination that the change is significant

The risk assessment should identify and analyze new risks, as well as changes to existing risks, that may result directly or indirectly from the proposed change.

A proposed change intended to improve the safety of a device may unintentionally introduce new risks or alter existing ones that could compromise the safety or effectiveness of the device. The risk assessment should therefore consider the potential for such unintended consequences.

Examples of possible unintended consequences that could be identified by a risk assessment:

• A delivery system is modified to prevent premature deployment, which unintentionally introduces a new feature or component that could damage or shorten the useful life of other components, or cause a failure itself.
  • For example, the device fails to deploy or only partially deploys.
• An alarm may be added to warn of an adverse event, but the sound or visual cue may resemble another alarm and thus may be misinterpreted. It may also trigger too frequently, leading to alarm fatigue and reducing the likelihood that users respond to higher-priority alarms.

A risk assessment may conclude that the change is likely non-significant if:

• no part of the change has been assessed as being significant and
• the probability of introducing unintended consequences is inherently negligible by design, based on the nature of the change

Using the four-corner principle

As a general principle, when a new modification to a device completely falls within the scope of existing licensed devices, the change is generally considered to be non-significant. This principle is based on the "four-corner testing" paradigm, which refers to:

• a method of assessing the extreme or most important aspects, features and characteristics of a system and
• how it performs or behaves in a variety of settings, situations or environments

In this paradigm, the 4 corners represent the most extreme conditions or features of a given system or of the environment in which it may be used.

For example, if a device is a tube available in different lengths and diameters, the 4 corners would be the:

1. longest length with the largest diameter
2. longest length with the smallest diameter
3. shortest length with the largest diameter
4. shortest length with the smallest diameter

Testing of these 4 corners would represent the extremes of the licensed models of the device. It would cover the full spectrum of expected performance for devices with lengths and diameters within the bounds defined by these four corners when used for the same indications for use.

The change is likely non-significant when:

• a proposed device modification does not introduce a feature or condition that exceeds the extremes of existing licensed devices or
• the extremes of the clinical environment in which they may be used

The change may be considered non-significant if the scope of the change:

• can be fully verified and validated based on test data that was previously submitted as part of a licence or licence amendment application and
• is fully applicable to the current licensed devices

Examples of a non-significant change:

• The length of the changed device is between the length of 2 existing licensed devices under the same licence, with all other features being the same.
  • You must verify that this change does not alter the safety or effectiveness of the device and that performance remains consistent with the existing range of devices.
• The indications for use have been narrowed to a subset of previously licensed and explicitly stated indications accepted during pre-market review of the subject devices.
  • This change must not have been made based on new data indicating the device is unsafe or ineffective for the previously licensed indications for use.

Examples of a significant change:

• The size of device is larger than the current range of licensed devices.
• The indications for use have been expanded to include a new clinical use case.

As an exception to this rule, if the behaviour of the device is difficult to predict based on the behaviour at the 4 corners, then the change is significant. This may occur:

• when performance could potentially be altered significantly based on minor changes within the 4 corners or
• when localized effects, such as a natural frequency effect, within the 4 corners can result in extreme outcomes

Examples of such a significant change:

• The size of the device is between 2 existing licensed device sizes where performance of the device is a non-linear or unpredictable within the 4 corners.
  • Such changes are significant because the performance of the device cannot be adequately predicted from existing data and device durability may be highly sensitive to changes in size.
• Magnetic resonance (MR) conditional claims have been modified or newly introduced for an active medical device.
  • Such changes are significant because the performance of these types of devices within an MR environment cannot be adequately predicted from existing data. This is because non-linear effects and natural frequency effects may occur within the MR environment.

Design verification and validation

Design verification testing ensures that a device meets its design specifications. Most device design changes will require design verification testing.

A change to a device is likely significant if design verification testing:

• produces unexpected results
  • in such cases, the risk assessment should be re-evaluated to determine if the change could affect the safety or effectiveness of the device
• differs from the verification testing conducted for a previous application, particularly when the change to the device is the reason new or modified testing is required
  • differences in testing may include changes to test methods, protocols or acceptance criteria

Design validation testing demonstrates that the device performs as intended clinically. The need for validation testing usually indicates that the change made to the device is significant. However, the absence of validation testing does not necessarily mean the change is non-significant.