We have been doing quite a lot of work for clients submitting their electrical devices for IEC 60601 Edition 3 electrical safety testing. For those unfamiliar, most users of medical devices would agree that mandatory electrical safety testing is a good thing before devices are allowed on the market. Now, 60601 Edition 3 takes this to a new level, the rights and wrongs of which are discussed elsewhere, as many of the risk management requirements specified in ISO14971 are reviewed as part of the test. Whether the intent of the IEC 60601-1 working committee was for compliance to be required for all devices currently marketed, or only new devices, is unknown, but Canada only requires 60601-1 Edition 3 for new devices marketed after 1 June 2012, whereas the deadline for ALL Devices in Europe was 1st June 2012.
The FDA required 3rd Edition for all new submissions and substantial changes to a device from 30th June 2013, but there is some confusion as to whether existing devices need to be re-certified. Elsewhere in the US, whether Testing Labs such as UL will continue to allow their logo to be affixed to devices without 3rd edition testing, or whether Hospital Bio-Medical Electronics equipment service departments will mandate 3rd edition, remains to be seen.
Over the course of the past year we have worked on several projects, and I wanted to relay our experience and processes we developed. Generally, if the product is ‘new’, i.e. less than 5 years old, and especially if already CE marked under the MDD (which requires ISO14971 risk management), then things were in generally good shape. As said, 60601 Ed 3 requires risk management to be considered in addition to the electrical safety testing. We found relatively new designs already have a comprehensive risk management file in line with ISO14971, and it was just a matter of some tweaking – IEC TR 62348 provides a cross-reference which allows a gap analysis to be undertaken, with test house checklists also being very useful.
For legacy (10 year old) devices which have been on the market for many years, things are a little more complicated. We found that, apart from risk management, usability (collateral standard 60601-1-6) and software validation has generally been weakest, with ‘Essential Performance’ requiring particular work due to expanded requirements in 3rd edition. Essential performance is often defined in the IEC 60601 collateral standards for a particular device class, but if not, then the manufacturer must state what the essential performance is. As an example, any critical alarm is part of Essential Performance. We also found that, unless a critical issue regarding patient safety or Essential Performance was involved, the risk management file often infrequently updated with post market surveillance information.
We generally found scarce or poorly defined input documentation like requirements specification, but good output documents like manufacturing drawings. In these cases, we found it was actually more efficient to develop documentation from scratch, rather than ‘force’ existing documentation – from this, here are my top tips for preparing documentation for 3rd edition 60601 testing for Legacy products:-
- Read the Standards! It is surprising how many people developing medical device have never looked at the appropriate standards. Although dry, 60601-1 3rd edition, collateral standards and usability/software standards can be skimmed through quite quickly.
- Create a Document Hierarchy Map in the Design or Quality Plan which clearly shows the relationship between documents – this aids the Test House reviewer, as putting them in a bad mood is ill-advised. An example is below, but software may require more documents depending on device class.
Figure 1. Typical Document Map for Legacy Device Submission. Red boxes are verification/validation, Green boxes are software, Dark Blue is usability, Light blue are input documents.
- A detailed review of the Post-Market surveillance data e.g. complaints, returns, MAUDE database, helps identify if there are any unaddressed issues (particularly usability) with the device, identifying showstopper that require rework.
- A detailed review of software in line with IEC 62304: ‘Medical device software – Software life cycle processes’ and/or FDA Document ‘Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices’, is required to establish the depth of the software documentation required, as different device classes have different requirements. Minor updates to software might be necessary in order to provide appropriate instrumentation required for unit testing. Also, legacy devices sometimes have software doing things which are now frowned upon. In these cases, it is always better to perform a design tweak; however, if the risk is low, we have been successful in demonstrating historical data shows a particular software failure mode is improbable, hence the risk is acceptable.
- Review current manufacturing drawings to ensure that appropriate safety factors are in place. Although 60601 is perceived as an electrical standard, it has a lot to say about mechanical and structural integrity, and requires 4x safety factor for brittle parts and 2.5x elsewhere. Again, we have shown that 2.5x safety factor is acceptable for brittle parts where the risk is low and historical data demonstrates no failures.
- Technical Specification should only contain measurable/testable items that are critical for Essential Performance or Safety. Otherwise, subsequent verification can be unnecessary and time consuming e.g. the colour is blue or the presence of a manufacturer logo. One of the reasons for creating new documentation is that legacy documentation often contains a single input document, which combines usability, business/marketing requirements and technical specifications, which should be relocated to the appropriate requirements document., and addressed in architecture.
- Documenting Usability Validation can be achieved by performing literature review of clinical papers, review of FDA Maude database and post-market complaints, and by actors emulating specific use cases called out by both the usability Requirements and Software usability. Often, manufacturers have close contacts with clinicians, so a minuted focus group meeting can also be used as evidence in the validation report.
Finally, this process can reveal issues about the device which may not be welcome by some. However, I view these opportunities for making the device ‘better’ – for example, the safety factors might work at 2.5x, but on the next spin of the device, 4x safety factor should be incorporated into the design. IEC 60601-1 3rd edition was actually first published in 2005, and will probably be updated in the future with even stricter requirements, so it is better to identify any issues sooner rather than later, and plan to upgrade the device as part of continuous improvement.