By Jake Rossen
By Jesse Marx
By Michelle LeBow
By Alleen Brown
By Maggie LaMaack
By CP Staff
By Jesse Marx
Enrollment in the study was cut short. By the time 800 participants, half of them Silzone users, had signed up, researchers had identified a high rate of paravalvular leak. (As it happened, the Mayo Clinic surgeon who removed Hatch's valve was one of the survey's lead investigators.) They stopped enrolling new patients. St. Jude immediately recalled the valve. In its communications with doctors, the company explained that the panel monitoring AVERT met twice in 1999 to discuss Butchart's findings about clots in British patients. They concluded that the findings "were not statistically significant."
In a written comment submitted for this story, St. Jude points out that thromboembolic events (clotting, that is) are a known complication of all heart valve replacement surgery, regardless the type of prosthetic valve used. "Mr. Butchart's experience with thromboembolic events was reported to St. Jude Medical in late 1998," writes Angela Craig, vice president for corporate relations. "St. Jude Medical thoroughly investigated Mr. Butchart's concerns and the results of this investigation were presented to regulatory agencies worldwide, including regulatory agencies in the United Kingdom and the United States (FDA). The experience reported by Mr. Butchart differed markedly from other centers using the Silzone valve and from that in the AVERT clinical trial, which was a large, multi-centered randomized clinical study which included the Silzone valve.
"The investigators in the AVERT study continue to follow Silzone patients to this day," Craig adds. "The latest data show that there is no evidence of an increased risk of complications for those currently implanted with a Silzone valve versus those implanted with a non-Silzone valve."
Many of the documents filed in U.S. District Court in Minnesota, where jurisdiction over most Silzone-related suits has been transferred, have been sealed because St. Jude deems them "proprietary." Portions of depositions and transcripts that are public suggest that when the cases finally go to trial, there may be further argument about what St. Jude should have known, and how early. Documents on file in federal courts indicate, for instance, that one of five sheep in one of the preliminary trials was found dead with a large clot on its valve, and that this sheep was then eliminated from the study, meaning the FDA never learned of the incident.
Other documents suggest that St. Jude sponsored another investigation, the "top accounts" study, but never publicized the results. Portions of a plaintiffs' deposition on file with the court quote e-mails between a biostatistician who reviewed this study and other researchers noting that thrombosis rates in its data were in the range of Butchart's controversial U.K. findings.
In a report defending the differences between his findings and St. Jude's, Butchart says that after his examination of the AVERT data he can't be sure that reports of clots were even logged prior to the year 2000.
The way Tom Hatch sees it, the debate is no substitute for the thorough pre-market trial that the FDA did not require St. Jude to perform. "I feel like I was their guinea pig to some extent and that it was a failed experiment," he says. "But I paid for it."
According to congressional investigators, some 15,000 companies market nearly 100,000 different medical-device product lines each year. The vast majority of these do not, like Tom Hatch's shiny new valve, have life-or-death powers. As defined by the FDA, medical devices include everything from relatively uncomplicated products like tongue depressors and home pregnancy tests to implantable devices like artificial hips, and life-sustaining things like pacemakers and coronary artery stents.
There were far fewer devices in 1976 when the FDA began regulating the industry, and most were much simpler. The new regulations required manufacturers to provide research showing that new devices were safe and effective, but devices already on the market were grandfathered in. And in an effort to reduce the red tape keeping worthy innovations off the market, Congress agreed that new devices that were "substantially equivalent" to existing, grandfathered-in ones could enjoy a streamlined approval process. The products subject to this fast-track approval have come to be known as 510(k)s, after the statute that spells out this exception in the law.
"You can understand if you have a bandage that's flesh-colored and a bandage that has Mickey Mouse on it, those are substantially equivalent," explains Peter Lurie, deputy director of Public Citizen's Health Research Group and a physician. "No one would be particularly interested in seeing that go through a more elaborate approval process."
But Congress meant for the FDA to scrutinize the most critical category of devices, those that sustained life or could kill the bearer, he adds. "There just really isn't much scrutiny that applies at all to those 510(k)s," says Lurie. "The very basic problem is that they're dealing with a very large volume of material that comes through there. My general impression is that [the] process is being used far more often than Congress contemplated. There is discretion left to the agency about how to assign products, but the way it's been done, the 510(k) has become more the rule than the exception."
The Silzone's basic architecture was that of the most popular heart valve in existence, the St. Jude mechanical valve, which had proven safe and effective for decades, and which is still used quite successfully. Based on experiments run in labs, the company thought that coating the valve's polyester sewing cuff with elemental silver might reduce further the thorniest complication from valve replacement surgery, infection.