Bench testing important for patient safety, successful device development

Although not tightly regulated, bench testing is essential for device development.

The bench testing of cardiovascular devices, particularly complex electronic devices that remain in patients for years, presents challenges for developers, industry and regulators.

Bench testing — a crucial step in the early device design process — is designed to tease out mechanical and design flaws in devices, and to test endurance of the device in the human body without having to implant the device in a human. Clinical studies on bench testing, whether on the actual results of the preliminary testing or in the preclinical results of tests in surrogates such as animals, are limited.

William H. Maisel, MD, assistant professor of medicine at Harvard Medical School and a cardiologist at Beth Israel Deaconess Medical Center in Boston, attributed this in part to individualized variations in protocol from different manufacturers producing differing devices.

“There are international standards and standard committees that try to level the playing field so that industry as a whole is doing bench testing in similar fashion with well-described tests,” Maisel told Cardiology Today. “The problem is that those are often consensus standards, so there have not been published studies describing bench testing and how it correlates, for example, with clinical performance.”

Approval processes

The types of bench tests employed depend on the type of device being tested. Currently, the FDA classifies and regulates the premarket approval of devices based on their potential risk to patients. According to the Advanced Medical Technology Association, class I devices such as tongue depressors and thermometers that present little risk to patients make up about 30% of all medical devices. Class II devices such as catheters and patient monitors represent about 60% of devices, and class III “breakthrough” devices such as artificial heart valves that present high risks to patients represent the remaining 10% of manufactured devices.

William H. Maisel, MD, said there have not been published studies describing bench testing. Source: Graham Gordon Ramsay

The most rigorous standards are reserved for class III devices, and the FDA regulations for these devices cover every phase of development from bench testing to the premarket trials and postmarketing follow-up. No regulations, however, dictate which specific tests are to occur during bench testing, regardless of the device.

Maisel told Cardiology Today that it could ultimately be in the interest of manufacturers to create industry-wide standards for bench testing and reporting of the test results.

“Although many manufacturers will not publicly state this, there is an advantage to having a better testing process, so companies don’t necessarily readily share what they learned in bench testing,” Maisel said. “If there was one manufacturer that developed an effective test at detecting a particular defect, that test would ideally be shared with the industry so that everyone could use that test and all devices would get better, but that is not how it really happens. There is competitive advantage for a given company with its individual testing.”

Recommendations

Although it does not intervene heavily regarding the type of bench testing manufacturers choose, the FDA requires that some devices demonstrate “substantial equivalence” to previously approved devices. In the case of metal expandable biliary stents, a 1998 guidance document from the FDA recommended deployment testing, expansion and compression force testing, dimension testing, corrosion testing, deformation testing, balloon expansion testing and tensile strength testing, as well as recommending the testing of at least three devices within each test. The specifics of each test, however, were not delineated and were left up to the manufacturer.

“If you are going to implant a device in someone that is going to be planted in a blood vessel that has contraction and relaxation, you could do rapid fatigue testing with 400 million repetitive back-and- forth movements with the device to make sure it doesn’t break,” Maisel said. “It can be an iterative process in the design and development stage.”

The results of such testing are often not shared explicitly with regulators or the public during the development and prototype testing phases before undergoing the FDA approval process, but Maisel noted this was common.

“A lot of that testing never sees the light of day, and a consumer or FDA reporter really does not need to see all of that development stage early on,” Maisel said. “A lot of that happens behind the scenes at a manufacturing plant or in a company where we do not hear about it, and that’s fine.”

Replicating the human body

Products that complete the premarket approval process mandated by FDA regulations may still malfunction when implanted in patients, sometimes years later. Maisel published a 2008 editorial in The New England Journal of Medicine noting that Medtronic’s Sprint Fidelis defibrillator leads had been approved by the FDA in 2004 despite the absence of clinical evidence of their performance in humans. In October 2007, Medtronic issued a voluntary recall of certain models of Sprint Fidelis leads due to reports of fractures in some patients. Despite going through extensive bench testing and completing the FDA premarket approval process for a class III device, weaknesses in some of the leads were not apparent until after the device had been on the market and implanted in patients.

Unforeseen malfunctions can be attributed in part to the unique and challenging environment of the human body.

“All three of the main device manufacturers — Boston Scientific, St. Jude and Medtronic — do extensive bench testing with all of their devices, including for pacemakers, defibrillators, catheters, ablation devices and so on,” Douglas P. Zipes, MD, distinguished professor of medicine at the Indiana University of School of Medicine in Indianapolis and section editor of the arrhythmia section of Cardiology Today, said in an interview. “None of the bench testing, however, can accurately replicate what happens in the human body, which is a very foreign and challenging environment to put any kind of equipment in and have it function for multiple years.”

Zipes and Maisel both agreed that rigorous bench testing is ultimately in the best interest of the manufacturer and, ultimately, the patients.

“Negative bench testing results are informative in the sense that if you find something not working during bench testing, you can be well assured that it is certainly not going to work inside the body,” Zipes said. “In contrast, if it works well during bench testing, obviously that is reassuring, but it is still no guarantee that it [will] work inside the body. The Fidelis leads are examples of that.” – by Eric Raible