Issue 47 / 3 December 2018

THE past week has seen the International Consortium of Investigative Journalists (ICIJ) release its Implant Filesthe result of an extensive investigation into the medical implant and device industry. Despite some enlightening reporting, the issues raised can be challenging to understand without a wider background in medical technology.

What is a medical implant?

For the most part, the ICIJ investigation has focused on “medical implants” — devices placed inside a patient to serve a specific medical function. Some of these implants can be very simple, such as a wire or a screw to hold broken bones together, a new hip or knee joint, or a catheter used for kidney dialysis or chemotherapy delivery — indeed even a suture used to sew up a wound (or a mesh used to reinforce and strengthen a wound) can be considered to be a medical implant.

Sometimes they are very complicated electronic and mechanical devices, such as pacemaker and defibrillators, spinal stimulators for chronic pain, computer-controlled medication pumps for diabetes or even left-ventricular assist devices, which act as an “external heart” until a patient can have a heart transplant.

In reality, however, implants are only a very small example of the medical technology that is all around us. Many medical devices are not implanted into patients — they may be tools that are used during surgery, such as endoscopes to look inside the bowel, retractors to hold organs out of the way, or staplers that allow surgeons to close wounds faster or repair and join bowel in tight spaces. They may be x-rays, or ultrasound machines that allow us to perform minimally invasive heart surgery through tiny punctures in the skin, or monitors that allow us to measure the depth of anaesthesia, nerve function, or oxygen levels.

Why does it matter? What is the risk?

We regulate medical devices for the same reason we regulate medication and pharmaceuticals. While there is great benefit from the use of these devices, they also come with risks, especially if used incorrectly.

For example, paracetamol (acetaminophen) is an extremely safe and effective drug sold over the counter in every drug store in the world. If misused, it can cause liver failure and death. There is therefore a moral (and usually a legal) responsibility for the manufacturer, vendor and dispenser to minimise the risk of harm through restriction of supply, responsible marketing to avoid promoting misuse, good design to discourage errors, and clear instructions to suppliers and consumers on how to use it safely.

The same responsibilities apply to any other medication, tool, device or implant that claims to have medical or health care benefits, and this forms the basis of the modern pharmaceutical and medical device regulation system. In Australia, this is managed by the Therapeutic Goods Administration (TGA). In North America, the responsible bodies are the US Food and Drug Administration (FDA) and Health Canada. In Europe, the European Medicines Agency coordinates drug regulation, while individual national agencies and private companies check for conformance (issuing the ubiquitous “CE Mark”) and will administer the new EU Medical Devices Regulation (MDR). Because of the increasing focus on risks related to medical devices, the International Medical Device Regulators Forum was established in 2011 to harmonise and coordinate systems worldwide.

The need for good medical device regulation has become increasingly evident, as the consequences of device failure have become more severe and more costly for patients, medical device manufacturers, and the community.

This has been exemplified by the increasing amount of litigation and cost of recalls from pelvic mesh surgery, PIP (Poly Implant Prothèse) breast implants, and Johnson and Johnson’s metal-on-metal hip implants over recent years.

How does device regulation work?

It is not practical to identify, list, register, monitor and test every drug, tool, instrument and device that is claimed to have a health care benefit; otherwise, regulators such as the TGA would have to test every new food that claims to make you feel better, lose weight, or have a positive outlook on life. Instead, most regulators take a risk-based approach.

If the drug, medicine or device does not make claims that are dangerously misleading (such as by telling people that it cures cancer), or, by its nature, it poses little or no material risk (such as crystal therapy), then there is little need to regulate or monitor it. This is how the Apple Watch ECG app was registered with minimal testing — it avoids making medical treatment claims and carries no theoretical risk of harm.

On the other hand, if the claims made are unproven and have a substantial risk of harming the public, then there is a need to have evidence that the claims are true and not misleading. Similarly, if the device is complicated, invasive, carries a risk of failure or misuse, or the consequences of failure or misuse are severe, then there is a clear need for systems to protect the public and minimise the risk of device failure or misuse, and to rescue any patients that are harmed by such failure.

Developing the evidence for bold claims through clinical trials is expensive and time consuming, as are quality control systems, failure-mode testing, training of health care providers, and monitoring systems. The onus is therefore on the manufacturer or supplier of the product to demonstrate its safety or risk profile — and to pay for the costs of doing so.

Agencies such as the TGA act purely to protect the public from harmful devices and claims, but make no judgement on how much they should cost or who should pay for them. That is a matter for individual hospitals, governments, and health insurers.

What are the problems with this system?

Profit-driven pharmaceutical companies and medical device manufacturers seeking to generate sales by convincing patients, doctors and regulators that there is a genuine need for their device, that the device solves the problem effectively, that it does not cause more harm than it solves, and that it is available for an acceptable cost. This profit motive inevitably provides an    incentive to selectively submit information, downplaying the negatives and exaggerating the positives.

Regulators must counter this by creating rules and systems that protect the public from harm, but these systems are expensive. In many regulatory models, the device manufacturer pays a fee that covers the cost of the regulatory agency’s expenses in checking and reviewing their data and claims, and this creates a potential conflict of interest. The TGA and similar agencies are almost wholly funded by fees charged to parties who are making submissions to them. The fees are high and provide a barrier to spurious submissions, but they are necessary in the absence of significant government funding.

Furthermore, they must balance the need to assess data and evidence from scratch and against local practice and conditions versus unnecessarily repeating research and studies, and delaying availability of a valuable treatment. As a result, agencies such as the TGA must take into account approval and submissions made to other regulatory agencies, including the FDA and the European CE Marking; however, this can lead to an expectation from patients and manufacturers of automatic approval once any international regulator has approved a technology, and places unnecessary pressure on the TGA.

We are moving towards an era when medical technology increasingly consists of things that are hard to measure or test. New devices and diagnostic software based on artificial intelligence or machine learning are coming, but while we all clamour for the promise of better cancer detection or magnetic resonance imaging scans, it is difficult to imagine how to regulate or control a self-learning computer algorithm. With 3D printing and personalised medicine, it becomes pointless to perform fatigue-testing on a device, such as a personalised aortic stent-graft or 3d-printed jaw, of which only one or two will ever be manufactured, and, therefore, regulation has to move towards process-based assessment rather than product-based.

Off-label use, adverse events and post-marketing surveillance

Once a medication, device or technology has been approved, it is able to be sold and marketed in Australia; however, this is only for the application for which the submission was originally made. There is little to stop a doctor using that medication or device in a manner or technique that is different to what was originally approved.

Such off-label use requires a doctor to disclose to the patient that the device is being used differently. Many modern advances in medical care – such as intravenous thrombolysis for stroke – are possible only because somebody took an existing medicine or equipment with which they were very familiar, had plenty of experience with, and tried something slightly different — generally in the setting of institutional review and oversight – and there are many commonplace treatments – such as ultrasound guided foam sclerotherapy – that  are routinely off-label because nobody will pay for a repeat application, even if it is guaranteed to be approved.

Often the evidence and data submitted to regulatory agencies only reflect the tightly controlled scenarios of clinical trials in which only patients in certain age groups and meeting strict criteria take part. Ongoing safety systems require post-marketing surveillance, in which device suppliers must proactively monitor problems that may appear in a real-world environment and take action to investigate and rectify them. To do this, they must maintain a close relationship with the hospitals where their devices are used and the doctors who use them.

Suppliers must update doctors with any changes to the instructions for use, follow-up guidelines and surveillance protocols for potential device failures, and rectification measures if problems arise during the use or follow-up of these devices. Every Australian hospital must maintain a log of every device used and in which patient, including the manufacture date, batch, lot and serial number of every medical device or implant. Many specialty-specific audit programs, such as the Australian Orthopaedic Association’s National Joint Replacement Registry or the Australian Breast Device Registry provide a third-party tracking of implant performance and long term outcomes, which is fed back to regulators and industry. These registers and databases have played important roles in detecting device-specific failures in the past.

As part of this process, device suppliers and manufacturers must collect and act on any report of an adverse event, which means an unexpected or undesirable outcome related to the use of a device.

These adverse events are usually reported directly to the company by the doctor or hospital, and rarely directly to the regulator. The ICIJ investigation fairly raises the issue that there is probably significant under-reporting of adverse events that could possibly be related to a device, and whether they should routinely be made directly to the TGA rather than to the company.

There is some criticism that patients are not routinely notified when a product advisory is issued that could be relevant to their implant. In most cases, they continue with regular check-ups, and action is only taken if a problem arises, in the same way that it would have been done had there not been a product advisory. In these cases, it could cause more harm to call and scare patients unnecessarily for a problem not relevant to them. It is exceedingly rare that a critical advisory notice requires patients to be tracked and recalled when they would not have otherwise been checked, but hospitals are very prepared to do this if necessary and have done so in the past.

Where to from here

Doctors can become the meat in the sandwich between patients demanding the latest and greatest technology, companies trying to build market share, regulators tasked with protecting the public but given few resources to do so, and hospitals and insurers who want to drive down the cost of these devices.

The Australian health system is unique and that means that results from clinical trials performed overseas do not necessarily apply here. There needs to be an increased culture of local testing and clinical trials for medical devices. This needs to be balanced against the costs of regulation and inappropriate delay to market.

Nevertheless, there is great potential for an effective and efficient regulatory system to drive Australasian innovation in biomedical technology and encourage a home-grown medical technology industry. It is incumbent on the medical profession to take a proactive role in regulating the regulators.

Dr Jason Chuen is Director of Vascular Surgery and the Medical 3D Printing Lab at Austin Health, and a Senior Fellow at the University of Melbourne. He is also a guest lecturer on new technology and surgery for the University of Sydney and has participated in consultations with the TGA. He has previously received honoraria, consulting fees, educational and research support funds from Abbott, Bard Medical, Boston Scientific, Cook Medical, Cordis, Endologix, Getinge, Medtronic, WL Gore and Vascutek Terumo.

 

The statements or opinions expressed in this article reflect the views of the authors and do not represent the official policy of the AMA, the MJA or MJA InSight unless that is so stated.

3 thoughts on “Medical devices and implant regulation: a primer

  1. Anonymous says:

    The logging of fracture fixation implants is a vexed question as previous attempts to ensure individually packaged and presented screws and wires was a disaster for the operating surgeon trying to manage difficult proceedures (particularly in the middle of the night with half trained staff!). This issue is raising its head again via the sterilisation guidelines despite reaction from orthopaedic surgeons who have to use these devices;bureaucracy gone mad/run by committees of non-users of the products affected.

  2. Anonymous says:

    we definitely need a registry of implanted mesh for hernia repair. it should be funded by industry themselves, early warning will save very costly litigation later on

  3. Anonymous says:

    The problem here is that prosthesis are believed to be the source of pain! It needs to be understood that prostheses do not have a nerve supply. Pain of neuralgic type occurs after surgery whether prostheses are used or not.

Leave a Reply

Your email address will not be published.