Originally Published EMDM May/June 2005
15th Year Anniversary
15 Years of Medical Device Directives
Dario Pirovano
Regulatory Affairs Director, EUCOMED
First implemented in 1990, the “new-approach” medical device directives have performed exceptionally well overall. So why are some calling for an overhaul of the regulatory system?
 |
On 20 June 1990, European Council president D. J. O’Malley signed the first new-generation medical device directive: 90/385/EEC on Active Implantable Medical Devices. This was one of three directives (originally there were supposed to be four) intended to harmonize regulations involving the placement of a medical device on the market and putting it into service throughout the European Union.
The Council Resolution of 7 May 1985 outlined the new approach to technical harmonization and standardization, giving legislators the tools to draft appropriate legislation for a sector characterized by an incredible rate of innovation.
The first draft of 90/385/EEC was presented to the various stakeholders in late 1988. It generated great interest among industry as well as authorities in the 12 member states. Industry immediately identified the measure’s potential to foster technical development under appropriate third-party control.
For some member states, 90/385/EEC was their first brush with medical device regulation. Some states wondered why medical devices had to be treated differently from pharmaceuticals. Those countries that already had some form of medical device regulation in place scrutinized the proposal to evaluate whether their existing rules were appropriately taken into account.
The watchword in those days at the Commission was, let’s not try to reinvent the wheel. New European legislation should be built on the national laws and regulatory experience of member states rather than invented from scratch or imported from abroad (such as US FDA procedures, for example).
The “new approach” was the right tool at the right time for putting this philosophy into practice in the European Union.
In the Beginning
In the second half of the 1980s, only four EU countries had specific medical device legislation and regulations in place: France, Germany, Italy, and the United Kingdom. The others had varying requirements, particularly for sterile devices, most of which were based on an existing pharmaceutical regimen.
France required formal product approval and subsequent registration (homologation) after a set of tests, mostly based on French standards, coupled with some sort of clinical evaluation. Clinical evaluations were performed by clinicians selected by the Ministry of Health (MoH) and focused on the user-friendliness of the products. Homologation was mandatory for a class of products named on a list that was constantly updated by the MoH.
Germany’s Medizine GeräteVerordnung (MedGeVe) required compulsory type testing of certain medical devices, mostly active (i.e., electrical) equipment, against German standards.
Italy’s system involved the registration of several identified devices (pacemakers, for example) and a more general registration of all plastic devices liable to come into contact with blood. Evaluations were performed by the Istituto Superiore di Sanità on the basis of its own testing protocols.
In the United Kingdom, the National Health System was encouraged to buy products only from manufacturers that had been registered by the Department of Health and Social Security (the grandfather of the Medicines and Healthcare Products Regulatory Agency). Manufacturers were evaluated based on their compliance with a set of guidance documents setting quality standards (the great-grandfather of ISO 13485).
If we compare these national systems with the structure of the medical device directives, it becomes apparent that the legislators laid out some basic principles:
• National systems are all appropriate, in spite of their diversity, to ensure the safety and performance of devices.
• If different systems are all appropriate, then the end result is equivalent. Therefore, the choice of a method showing compliance should be left up to the manufacturers, who can make an informed decision based on the characteristics of their business activity.
A System That Works . . .
During the past 15 years, the medical device directive system has contributed to improving patient access to modern, innovative, and reliable medical technology. This has been achieved by:
• Formally binding manufacturers to declare that they have complied with the relevant requirements and that consequently their devices show a positive benefit-to-risk ratio when they are used as intended, and that they effectively perform as intended.
• Imposing the conformity of all medical devices to a set of essential requirements designed to address safety and performance.
• Generalizing the use of quality systems.
• Putting in place an effective vigilance system.
• Establishing a set of conformity assessment procedures, where the extent of third-party involvement is proportional to human vulnerability in case of failure.
• Giving manufacturers the possibility to choose from among various and equally rigorous conformity assessment procedures the one that is most appropriate.
. . . But Is It Perfect?
Perfect systems, like absolute safety, do not exist. The medical device directive system is not an exception. In 2002, the European Commission and Council published a report on the functioning of the Medical Devices Directive (MDD). Three basic concepts were highlighted:
• The MDD is an adequate tool for ensuring the safety and performance of medical products.
• Overall, the potential deficiencies do not stem from the actual requirements but, rather, from their implementation.
• Some requirements, namely in the areas of conformity assessment based on quality systems and in clinical evaluations, need clarification to harmonize the quality of the controls.
I believe that the first point demands particular attention. Ever since the political establishment discovered the medical device sector—in some member states, public expenditures for medical devices surpass drugs—a debate has flared, questioning why the system applied to pharmaceuticals, perceived as being “more rigorous,” is not applied to medical devices. Medical equipment is covered only by the CE mark “which applies also to toys.” (This surprising declaration was made in October by a member of the Italian Parliament during a press conference in Rome.)
It is extremely important to clarify this point and to broadcast the message to the public. The system for pharmaceuticals was developed more than 40 years ago to regulate products that:
• Share a similar industrial profile (basically a chemical process).
• Are not likely to change over the years.
• Are sold in extremely large quantities.
• Have relatively low production costs (in terms of materials).
• Generally cannot rely on in vitro tests to provide conclusive evidence of efficacy and safety.
Conversely, the medical technology sector is made up of companies that:
• Operate in virtually all of the industrial and scientific branches, from microbiology to mechanical engineering.
• May develop products with a life cycle as short as one year.
• Generally do not sell huge quantities of products.
• Have intrinsic production costs that are, in many cases, higher than those for an average pharmaceutical product.
• Often can use in vitro tests to verify product safety and performance.
This comparison makes it immediately apparent that the more rigorous pharmaceutical system is not necessarily the most appropriate for medical devices.
It should also be noted that toys are CE marked purely on the basis of a manufacturer’s declaration, whereas medical devices are CE marked after having been subjected to a third-party evaluation proportional to human vulnerability in case of failure. CE marking requirements are not identical for all products covered by the new-approach directives.
How Competent Are the Authorities?
In some cases, greater involvement of the competent authority is called for during the premarket phase. It is important to note that a premarket assessment only makes sense if the assessor’s knowledge of the specific product is at least as good as the manufacturer’s. In Europe, no competent authority can claim such wide-ranging expertise for all products covered by the MDD. Notified bodies, which have more-flexible access to experts, cannot be replaced by competent authorities. Even the mighty FDA cannot claim to be an expert in all domains, even with a budget that is several times larger than that of the largest competent authority in Europe. But that is another story.
Which brings us to the second issue raised in the European Commission and Council report: implementation. As previously noted, competent authorities have neither the budget nor the personnel to perform the duties of notified bodies. Their responsibility is to evaluate the entities that apply to become notified bodies and to monitor their performance. Competent authorities are also deemed to put in place effective market surveillance. Generally speaking, this initiative has not been successful. It is not surprising, therefore, that the European Commission and member states have delegated two working groups—the Notified Bodies Operation Group and Market Surveillance Operation Group—to develop better guidance.
The last item in the report, involving quality systems and clinical trials, in theory is seen as a departure. In reality, the proposed modifications to the MDD simply clarify issues that, in my opinion, were already embedded in the original text.
Quality systems are based on three pillars: you must have a procedure (related to design, for example); the procedure must be appropriate; and the procedure must be followed.
Clearly, if you have chosen to apply a quality system as a route to conformity assessment, the notified body can perform an audit by evaluating a representative sample of the design files, regardless of the class of product. This point, apparently, has not been widely understood. The revision of Directive 93/42/EEC will take this into account almost word for word.
The need to perform clinical investigations has never been limited to Class III and implantable devices, simply because the need for clinical investigations is not a stand-alone requirement. It is the result of an evaluation to determine the appropriate method to show conformity to a specific essential requirement. It seems that this concept will be introduced using different words, but that revision will not change the practices of those device manufacturers who correctly applied the directive in the first place.
Drugs versus Devices
One disturbing note, however, is the encroachment of Directive 2001/20/EC. According to the Commission, this directive only applies to medicinal products, but some member states have a different interpretation. Clinical investigations involving devices are very different from those involving medicinal products. Trying to use the methods of the latter for evaluating medical devices makes as much sense as trying to measure time in centimetres.
Consider the concept of comparison. While it is acceptable that a new drug is made available to the public only if can be shown to have a better efficacy profile than an existing product, determining the appropriate performance of a medical device is linked to several factors, such as the population to be served, the average frequency of use of all of the device features, and so forth.
For example, some sophisticated pacemakers can treat virtually all forms of arrhythmia, depending on the way they are programmed. Others are designed for specific pathologies. If the practitioner has a clear idea of the patient’s disease and its development, he or she will be happy to use one of the latter types and reserve the most sophisticated device to treat a more uncertain condition.
To cite another example, MRI scanners provide state-of-the-art imaging for certain pathologies, but an x-ray machine is sufficient to check if a patient has broken
a leg while skiing.
The MDD is a safeguard against what I call the Windows syndrome, a disease that compels you to buy the most recent versions of software, when, in fact, you will only use 20% of its capabilities (the same ones you used in the old version).
Those who question the appropriateness of the Medical Devices Directive need to know that a CE-marked medical device is as safe as one that has been evaluated by US FDA. It is surely safer than a medical device that has been evaluated using a regulatory system conceived for an entirely different type of product.
Dario Pirovano, PhD, holds a degree in mechanical engineering with a specialization in medical technologies. At the European Commission, he played a key role in the drafting of Directives 90/385/EEC and 93/42/EEC. A consultant on regulatory matters for more than 10 years, he has lent his expertise to manufacturers, notified bodies, and competent authorities. Since 2002, he has served as a consultant on a quasiexclusive basis with EUCOMED, where he is responsible for regulatory affairs.
Copyright ©2005 European Medical Device Manufacturer
