With Medical Implants, Hope and Plan for the Best, but Talk About the Worst

The general public’s perception that surgery can fix anything is right sometimes: An innovative surgical technique, paired with a machine-learning interface, is allowing amputee patients at the University of Michigan Health System to control upper-limb prostheses with startling dexterity.¹ But outcomes aren’t always so positive. A recent article in The New Yorker began: “Do Brain Implants Change Your Identity? As neural devices proliferate, so do reports of personality changes, foundering relationships, and people who want to leave their careers.”²

Most of our concerns about combining foreign bodies with the human body aren’t as wild as personality change, but such headlines remind us of the uncomfortable truth that when we combine implants and humans, the results are unpredictable. Despite our best efforts at research, training, testing, and planning, variability among human systems leaves us with an irreducible degree of uncertainty when approaching many medical procedures.

As various implant-related advances are improving quality of life for type 1 diabetes patients³ and the perennial search for an artificial heart has shown renewed promise to extend lives,⁴ we can see that, with so many exciting developments underway, we don’t want to limit our options for helping patients. But we also don’t want to overpromise to patients who may not understand risks—unless we participate in a true informed consent process using shared decision making.

The Power of Surgical Advances Paired With Machine Learning

Paul Cederna, MD, Professor of Plastic Surgery at the University of Michigan, and Cynthia Chestek, PhD, Associate Professor of Biomedical Engineering at the University of Michigan College of Engineering, have combined a pioneering surgical technique with machine learning to solve a long-standing puzzle of prosthetic limb design. In order for the user to control the prosthesis with their mind, as they would their original hand, the device needs to be able to send and receive nerve signals—which are notoriously quiet, relative to signals from muscles and other local structures. At a recent conference hosted by the New York Regional Society of Plastic Surgeons, I was fortunate to hear Dr. Cederna speak about how the Michigan team has resolved this issue by wrapping nerve endings in small bundles of muscle tissue, thus amplifying the nerve’s signal.

These small muscle grafts, combined with machine learning, “provide individual finger control of prosthetic devices using the nerves in a patient’s residual limb,” says Dr. Cederna. “This opens up a whole new world for people who are upper-limb prosthesis users,” Dr. Chestek explains.¹ This particular innovation is so new that patients are not yet allowed to take their prostheses out of the lab. Patients are helping researchers test the level of dexterity they can achieve, so the range of individual experience with this intervention is still being discovered. Innovations like these are in development around the U.S. and the world. And while such unprecedented interventions come with risks, the benefits they could deliver are so impressive that it is easy to imagine how well-informed patients would choose to proceed, and how physicians would be excited to deliver this care.

The Risks of Not Explaining the Risks

Due to variations in how different individuals react to foreign bodies, we cannot completely eliminate or predict certain risks, even with well-studied procedures. Common procedures—such as intraocular lenses for cataracts, cardiac stents, and assorted orthopedic implants for trauma and arthritis—are not without implant-related complications.

These risks and uncertainties must be communicated to patients—who may believe that because a procedure is routine, it is foolproof. This obligation to participate in a true informed consent process is reflected in the FDA’s new rules for breast implants, which place less emphasis on changes in procedures and more on requirements for how the risks of those procedures are explained.⁵

When we do not communicate risks to a patient, it does not raise the probability that their system will mix badly with an implant, but it does raise the physician’s odds of being sued should the patient suffer an adverse event.⁶ Again and again, across specialties, we see that medical malpractice allegations are grounded in the communication as much as the care.⁷

The Essential Process of Shared Decision Making

The SHARE Approach, developed by the Agency for Healthcare Research and Quality (AHRQ), models a five-step process for shared decision making that includes exploring and comparing the benefits, harms, and risks of each treatment option through meaningful dialogue about what matters most to the patient. Crucially, one benefit of a shared decision-making approach is that it can help the physician determine the patient’s understanding of what has been discussed. As we all know, each patient’s level of medical literacy varies. When you combine that factor with their hopes and fears about potential procedures, plus potential cultural or language barriers and the ordinary fallibilities of human communication, the prospect of risk information being missed or misunderstood is constant.

Therefore, time spent accessing materials on shared decision making from AHRQ, or the materials that many medical specialty societies have created, is a wise investment. I also recommend The Doctors Company’s “Best Practices in Patient-Centered Care and Shared Decision Making.”

Advances in care are extending the length and transforming the quality of patients’ lives, so we should not let life’s uncertainties deter us from providing the best care we can imagine. But we should communicate risks with patients up front and base our informed consent process in shared decision making, so that we are confident that patients understand a procedure’s risks and they feel confident in choosing its benefits.