How to make state of the art medical devices

This is an older version, kept on my site because other articles link to it. I recommend switching to the newer version.
7 minute read.
Beginning in 2020 the new European Union Medical Device Regulation will protect patient safety by requiring healthcare companies to make products that are “state of the art,” a term that’s often misunderstood by both companies and patients. It does not mean the latest technology, it means the most commonly agreed upon technology, or “generally accepted state of the art.”
This European law is better explained through examples using automobiles, comparing today’s state of the art with new technology in the 1980’s television series Knight Rider, where David Hasselhoff fought crime in a talking, self-driving car.
Watching the 1-minute trailer may brighten your day:

Today, we have talking and self-driving cars, but they are not considered “state of the art” by government regulations, which require state of the art safety features. You’re probably familiar with generally accepted state of the art for automobile safety:
SeatbeltsBrake lightsAir bagsChild safety-seat securing hooksSide-impact safe doors
Self-driving cars exist but they are not mandatory because there’s not enough evidence that they protect public safety yet. Self-driving cars are a new, high-tech feature but are not “generally accepted state of the art.” Similarly, some medical device features seem like good ideas but aren’t “generally accepted state of the art” because there’s not enough evidence that they reduce risk to patients. The MDR requires companies to continuously review competition and public safety reports to prove that they’ve reduced risk to patients As Far As Possible, which includes having state of the art products.
This is easily understood with for cars; in 2016 an estimated 16,000 lives were saved in the U.S. thanks to seat belts. But, healthcare is a bigger risk than automobile accidents, with 80,000 – 250,000 unnecessary deaths in the U.S. each year due to healthcare errors. Similar data worldwide led to the EU MDR, and the U.S. is considering similar healthcare reform. But medical device safety is more complex than car safety, and state of the art is a combination of design, manufacturing process, surgeon training, etc. I’ll demonstrate this with a few examples.
Spine implants:
new technology, not state of the art
This is an example of a new medical technology that’s not state of the art: motion preserving spine implants.

Before motion-preserving spine implants, surgeons used spine-fusion implants to prevent spinal vertebra from moving, usually to protect the spinal cord, sometimes to reduce pain (this is a controversial topic – learn more.)
A concern developed that fusing one set of vertebra caused more motion in other vertebra which led to problems in 6 to 8 years. In the 2000’s several start-up companies developed “motion preserving” spine implants to replace fusion devices.

Motion-preserving devices have not become state of the art because the benefits are unproven and the new technology has higher risks than previous technologies. The added risk come from surgeons having to learn new procedures that are more complex than previous spine implants, and from unknown long-term consequences of the new technologies. The implants shifted over time and many patients suffered unnecessarily. Manufacturers settled class action lawsuits and national health insurance programs refuse to pay for the procedure, especially because it’s 10X more expensive than previous spine implants. Long-term studies of patients with motion-preserving implants haven’t shown benefits that justify the risks or costs.
In other words, new spine technology does not mean it’s state of the art healthcare, it could be adding both risk and costs to public healthcare.
Previous medical device regulations did not enforce state of the art safety, which is why products that add risk and cost are still on the market. Unfortunately, most patients rely on their physicians to advise them despite many surgeons are unaware of the risk/benefit analysis or cost. And, some physicians are incentivised by medical device companies to suggest the more expensive implants. All of this is why the MDR will hopefully benefit society.
Robotic Surgery:
beneficial, not state of the art

This is an example of even new, beneficial technology may not be state of the art because not all hospitals are capable of using the new technology yet.
Many spine surgeries use robotic surgery or some type of nerve-monitoring technology to protect patients’ spinal cords during surgery. This new technology is generally considered beneficial, therefore would almost be considered “state of the art” by the MDR definition, but it is not because it is only true in specific cases where the hospitals have sufficient infrastructure and surgeons are sufficiently trained and experienced. The EU MDR applies to all countries in the European Union, therefore to be state of the art the technology would have to apply to the infrastructure and training of almost 30 European countries.
In other words, a company can sell nerve-monitoring equipment to hospitals based on improved results, but not all spine implant companies would be held to the standard of robotic surgery outcomes because these new technologies aren’t yet “generally accepted state of the art.”
Hip implants:
established technology, not state of the art
Hip implants have been available since 1940, before David Hasselhoff was born, yet we still struggle understanding what is and what is not state of the art because designing, manufacturing, and shipping medical devices is much more complex than automobile safety features like seat belts and child safety seats.
To emphasize the complexity of state of the art for medical devices I’ll use terminology that probably seems confusing unless you’re familiar with hip implants, which is a challenge for patients having informed choices in healthcare.

When a patient needs a hip implant they rarely don’t review the current materials, noting differences between percentages of Chrome in CoChMg femoral heads, hardness and pitting resistance from heat-treating, or smoothness from polishing. We don’t know to investigate the density and cross-linking of UHMWPE in plastic liners, or the pore size, edge sharpness, and structure of nano-material porous structures on the acetabular component and femoral stem. Nor do we know if the coating is Ti6Al4V or Ti6Al4V ELI, much less know if their processing ensures that a titanium oxide layer forms and prevents the toxic material Vanadium from leaching out.
Are the the tools and instruments used for the surgery state of the art? Tools that are improperly categorized or instruments that aren’t user-friendly have led to mis-matched implants that failed and require secondary surgeries. Do the instruments ensure proper alignment of the hip stem? As little as 3 degrees variance can add risk of eventual failure.
Has the company’s quality system ensured replacing all instruments with updated versions? Are surgeon’s adequately trained, especially knowing that research shows that 60% of the reason for an implant failure is the skill of the surgeon. What if a surgeon has higher failure rates, is that tracked and monitored so a patient can make informed choices?
What about how the implant was made? We don’t know if manufacturing processes have state of the art cleaning procedures or if a company’s quality control is state of the art and would catch mistakes.

An example of a quality-system gap in hip implants is the Sulzer hip stem recall. 35,000 hip stems were shipped with toxic machining oil still in them. Over 9,000 were implanted before the mistake was caught, and almost 4,000 people had their femoral bones erode, requiring another surgery and impacting their ability to walk for the rest of their lives. Over $1 Billion in lawsuits bankrupted Sulzer, but patients said they’d rather walk normally than have received insurance money.
The reason for Sulzer’s recall was traced to a seemingly simple decision on their manufacturing line that even today would be difficult to detect and monitor, much less simplify for patients to understand. The Sulzer hip implant recall was just one example, many more exist.
Oh shit, now what?
The EU-MDR can’t define what is state of the art for every situation so it requires that each product be compared to competitive products’ safety features every year for Class III and IIb products, every 2 years for Class IIa, and “as needed” for Class I. This makes sense: the higher the Class, the higher the risk, therefore the more frequently the need to reduce risk. Class I products are low-risk, therefore “as needed” is ambiguous and implies if there’s a “major” design change or safety concern.
This means that “state of the art” will have two supporting documents, clinical data and post-market surveillance. Clinical data includes competitors’ products, safety data, and alternative treatments; post-market surveillance includes the manufacturer’s product safety. For now, companies must do their best to find this information publicly, but in the future all companies and the general public will have access to this information online. The MDR is creating new agencies that will look at this data and see if medical device features truly improve patient safety; if so, those features will become state of the art. This is similar to how governments currently treat automotive safety.
Government regulations require state of the art in new cars but do not enforce all innovations until there’s enough evidence that these features impact public safety. Agencies such as the U.S. National Highway Traffic Safety Administration (NHTSA) keep consumers informed about new technologies that aren’t required.

Over time, these features may prove that they add to public safety enough to be considered state of the art, at which time they may become required by regulations. Similarly, medical devices sold in Europe will be tracked online in the EUDAMED database, which is still being designed and will probably continuously improve.

The EU-MDR will try to minimize risks to patients by ensuring that new technologies are compared to generally accepted state of the art in terms of patient safety, and that new technologies are justified when the benefits outweigh additional risks. See my article on risk-benefit analysis for that step.
“State of the art” for the EU MDR does not mean the latest technology, it means the features and systems that are proven to reduce risk to patients.State of the art must be established every 1-2 years for Class IIa-III products and “as necessary” for Class III products.State of the art is complex, based on a combination of product features, manufacturing processes, training, and the realities of hospital systems in diverse countries.The MDR will create an online database, EUDAMED, with transparent, public data on medical device safety. That data will be used to determine “state of the art,” but until then companies must seek and use published research data.

State of the art requires complying with Risk Management priorities, which is described in another article Reducing Risk As Far As Possible
Learn more
There’s more to MDR. For example, there are requirements on how to update safety concerns and which information must be displayed on a company’s web page. You can learn in my articles on The Big Picture or, if you’re familiar with previous European medical directives, “MDR: the medical device regulation formerly known as MDD
The MDR will be mandatory by 2020; to be fully prepared consider working with one of these training or consulting companies to help your team prepare.
Oriel STAT-A-MATRIX an international organization since 1968 (I consult with Oriel)

MaetricsThe Weinberg Group

LNE group for consulting, and their subsidiary GMED for certification
Qunique, based in Switzerland and expanding internationally

Jason (me)

Thank you
Please share this article if you think others could benefit.
Follow me on Linkedin for occasional articles. Not all articles are published, so consider browsing my blog or subscribing for annual updates.
For fun…
Someone spent 10 years converting their 1984 Trans Am into a replica of David Hasselhoff’s Knight Rider car, complete with the same voice and 1980’s “state of the art” technology. Watch it here: