Fatal flaw: when medical devices fail, investigators connect the dots

When consumer technology fails, it's frustrating. When a medical device fails, the effects can be fatal. Meet the ECRI Institute, watchdog for gadgets in clinics and hospitals across the globe.
Written by Christina Hernandez Sherwood, Contributing Writer

By the time Mark Bruley got the call, the patient had already died on the table. A relatively straightforward surgery had gone terribly wrong. Surgeons were supposed to insert a blood clot filter, where clots from the legs are caught and dissolved before they reach the lungs.

But then the patient died. It was Bruley's job to find out why.

As head of accident and forensic investigations at ECRI Institute, a non-profit organization in Pennsylvania that researches the best approaches to patient care, Bruley serves as a sort of medical accident detective. Within weeks, he was at the hospital to investigate. He reviewed medical records and watched X-ray videos from the surgery. He watched as the filter, which looks like the ribs of an open umbrella, was correctly positioned in a vein between the heart and the lower body.

On to the next step. Bruley gathered with those involved -- the hospital, the manufacturer, the patient's family and their attorneys -- to inspect the filter itself.

Before heading to the meeting, Bruley checked the U.S. Food and Drug Administration's Manufacturer and User Facility Device Experience, a public database that aggregates reports of medical device problems. His screen populated with a handful of previous incidents. The problem, as reported: the capsule holding the filter detached from the delivery catheter because of contamination at the bonding site.

"All of those reports were before this fatal incident," Bruley said.

In the case of the patient who died, he said, the detached capsule flowed with the patient's blood into the heart. As the muscles of the heart contracted, the capsule perforated the organ, causing fatal bleeding. Days after the completion of a months-long joint investigation into the case, the manufacturer recalled about 18,000 filters worldwide.

Bruley and his colleagues are among the few people in the country -- even the world -- to independently investigate medical accidents and test health devices. In the U.S., the FDA relies on manufacturer's reports to evaluate medical devices. A recent investigation by Consumer Reports found that many medical implants aren't tested before use. Tucked in residential Plymouth Meeting, about 20 miles northwest of downtown Philadelphia, ECRI investigators research surgical fires, test patient beds and alert hospitals to health device hazards. With the motto "smarter choices, safer patients," ECRI keeps tabs on the technology used in hospitals around the world.

A Consumer Reports for medical devices

ECRI is one of the few evidence-based practice centers in North America and one of the very few not associated with a medical school or university, said Robert Logan, communications research scientist for the U.S. National Library of Medicine. "They really are a separate structure," he said. "That allows them to self-generate their own rules. Some of the guidelines they use to evaluate products are among the highest bar imaginable."

Joel Nobel founded the Emergency Care Research Institute in 1968 after watching a child die while a defibrillator failed. (The acronym ECRI was adopted in the early 1970s as the organization expanded its work.) The independent evaluation group struck out on its own when a medical journal refused to publish its first critical review of resuscitators. Modeling its process on that of the Consumers Union, which publishes Consumer Reports, ECRI's medical device evaluations now reach doctors, nurses, medical staff, engineers and others at 3,000 hospitals worldwide. Hospital members fund ECRI's work, and to maintain an extra layer of independence, ECRI employees are prohibited from owning stock in medical technology companies.

Even with more than four decades of testing under its belt, ECRI's work is constantly changing. James Keller Jr., who directs health technology evaluation and safety, says that ECRI is forced to keep up as medical tech evolves and manufacturers pump out more and increasingly complex devices.

Some devices make patients safer, like a system that helped one hospital discover an uptick in medical errors between 6 and 7 p.m. (To prevent further mishaps, the hospital rescheduled medical rounds to a less busy hour.) But technology doesn't always help, said Erin Sparnon, a senior project engineer at ECRI. A California hospital made news when it upgraded to a multi-million dollar physician ordering system, only to tear it out when the medical staff protested.

"Adding technology to a broken process," Sparnon said, "just means a faster broken process."

While ECRI often reviews "bread and butter" technologies such as defibrillators and patient monitors, investigators also look to the group's hospital members when deciding which devices to evaluate. Is a hospital on the receiving end of a manufacturer's sales pitch? Are hospitals buying up a hot new product? Is there controversy surrounding a technology?

Because medical devices can cost thousands, even millions, of dollars, ECRI receives products for review on loan. Investigators test each device based on a specific set of factors, like safety and integration, before publishing a product review in the monthly magazine Health Devices. "We often hear people talk about this as the Bible," Keller said, holding up a recent issue with intensive care ventilators on the cover.

If the evaluation uncovers a problem with the device, ECRI contacts the manufacturer immediately -- rather than waiting until the next issue of Health Devices. When the review is eventually published, ECRI mentions the initial device problem even if it's since been corrected. Over a recent five-year span, Keller said, 15 percent of product reviews in Health Devices noted an initial device problem ECRI identified that led to manufacturer improvement. "We feel that we're making a difference," he said.

It's through these reviews, Keller said, that ECRI helped drive the market toward safer infusion pumps, devices that deliver fluids to patients. After new pumps featured dose error reduction systems that warned medical providers of potentially dangerous mistakes, ECRI's investigators gave the devices preferred ratings. Now, products without the feature garner poor ratings from ECRI. "We really help to drive technology innovation," he said.

Detective work completes the safety picture

Five years ago, ECRI launched its annual Top 10 Health Technology Hazards list, intended to call out the biggest medical dangers in an effort to help hospitals set safety priorities. The criteria for determining what goes on the list: Can it kill or cause serious injury? How often do adverse events occur? How widespread is the technology? How high-profile is the problem? Some hazards, like surgical fires, show up on the list year after year. Often, problems stem not from faulty devices, but from the need for providers to use them more safely.

Bruley's office is down the hall from where ECRI investigators test patient beds and hospital alarms. As accident investigator, he spends much of his time traveling the world to visit hospitals that need an independent eye. When I visited ECRI last month, Bruley had just finished a 45-minute phone call with a hospital where a patient's head was set ablaze during surgery. He left almost immediately to investigate. "We basically run an emergency department for hospital-related accidents," Bruley said.

Bruley gets calls about surgical fires so often -- a couple of times a week -- that he's prepared a lecture for hospital staff members on preventing the incidents. Bruley even adorns his workspace with a poster showing him posing next to Smokey the Bear. The caption reads: "Only you can prevent surgical fires."

Postwar humor aside, it's a grim issue. One of Bruley's most well-known cases was a medical fire that gave a newborn severe burns. The baby, who had some health problems, was resting under a dome of oxygen in a bassinet below a radiant warmer. With a Kewpie doll as a prop, Bruley described the tragedy: oxygen released from under the dome had enriched the baby's blankets and diaper, causing the material to become ignitable. Overhead, the radiant warmer's spring-loaded heating element chipped, sending a spark into the bassinet. The subsequent burns on the baby's head, chest and arms were serious enough to require separate treatment at a burn center.

According to the FDA's device database, there had been several similar incidents related to the device -- and even a recall. But because the company changed hands, the new owners were unaware of the problem. Bruley's advice to the hospital: remove -- and destroy -- the infant radiant warmers in question.

Another Bruley investigation focused on a California clinic discovered to have contaminated endoscopes, which are used to examine the interior of the body. The clinic had switched to a new cleaning solution; staff members were told it was the same strength as the previous cleaner. But a simple review of the label revealed the disconnect: while the previous solution needed just 20 minutes of soaking time, the new cleaner required six hours.

Humans aren't perfect, and so neither are the medical devices they make. Given the breakneck speed of technological advancement, there remains a need for someone to connect the dots when something goes awry in a medical facility. Even if that means standing in the operating room while a surgeon does his work.

"We get involved to solve a problem and make recommendations," Bruley said. "[It's] all in the name of patient safety."

Photo based on original by Parker Michael White.

Editor's note: The original version of this story described the death of an infant from a surgical fire; that is incorrect. While the burns were indeed severe, the baby did not die. We regret the error.

This post was originally published on Smartplanet.com

Editorial standards