c.2013 New York Times News Service

c.2013 New York Times News Service

Doctors at Cleveland Clinic began to suspect in 2012 that something might be wrong with a high-tech implant used to treat patients with advanced heart failure like former Vice President Dick Cheney.

The number of patients developing potentially fatal blood clots soon after getting the implant seemed to be rising. Then early this year, researchers completed a check of hospital records and their concern turned to alarm.

The data showed that the incidence of blood clots among patients who got the device, called the HeartMate II, after March 2011 was nearly four times the incidence in patients who had gotten the same device in previous years. Patients who developed pump-related clots died or needed emergency steps like heart transplants or device replacements to save them.

“When we got the data, we said, ‘Wow,’” said Dr. Randall C. Starling, a cardiologist at Cleveland Clinic.

On Wednesday, The New England Journal of Medicine posted a study on its website detailing the findings from Cleveland Clinic and two other hospitals about the device. The HeartMate II belongs to a category of products known as a left ventricular assist device, and it contains a pump that continuously pushes blood through the heart.

The abrupt increase in pump-related blood clots reported in the study is likely to raise questions about whether its manufacturer, Thoratec Corp., modified the device, either intentionally or accidentally. By March, Cleveland Clinic had informed both Thoratec and the Food and Drug Administration about the problems seen there, Starling said.

Officials of Thoratec declined to be interviewed. But in a statement, the company, which is based in Pleasanton, Calif., said that the HeartMate II had been intensively studied and used in more 16,000 patients worldwide with excellent results. It added that the six-month survival rate of patients who receive the device had remained consistently high.

“Individual center experience with thrombosis varies significantly, and Thoratec actively partners with clinicians at all centers to minimize this risk,” the company said in a statement.

Thoratec and other cardiologists also pointed to a federally funded registry that shows a smaller rise in the rate of blood clots, or thrombosis, among patients getting a HeartMate II than the one reported Wednesday by the three hospitals. In the registry, which is known as Intermacs, the rate of pump-related blood clots associated with the HeartMate II rose to about 5 percent in devices implanted after May 2011 compared to about 2 percent in previous years.

The data reported on Wednesday in The New England Journal of Medicine found rates of clot formation two months after a device’s implant had risen to 8.4 percent after March 2011 from 2.2 percent in earlier years. Researchers also suggested in the study that the Intermacs registry might not capture all cases of pump-related blood clots, such as when patients gets emergency heart transplants after a clot forms.

Not only did the rate of blood clots increase, but the clots occurred much sooner than in the past, according to the study. After March 2011, the median time before a clot was 2.7 months, compared with 18.6 months in previous years. The report Wednesday also included data from Duke University and Washington University in St. Louis.

All mechanical heart implants are prone to producing blood clots that can form on a device’s surface. And experts say the rate of blood clot formation can be affected by a variety of factors such as changes in the use of blood-thinning drugs or the patient’s health.

In a telephone interview, Starling described Thoratec officials as cooperative, adding that they have been looking into the problem since March to understand its cause. He said that he could only speculate about the reason for the rapid rise in early blood clots but believed it was probably device-related.

“My belief is that it is something as subtle as a change in software that affects pump flow or heat dissipation near a bearing,” said Starling, who is a Thoratec consultant.

Asked about his comments, Thoratec said it had yet to determine the reason for even the smaller rise in blood clots seen in the federally funded database.

“We have performed extensive analysis on HeartMate II and have not identified any change that would cause the increase observed in the Intermacs registry,” the company said.

In a statement, the FDA said that it was reviewing the findings.

“The agency shares the authors concerns about the possibility of increased pump thrombosis,” the FDA said in a statement.

The fortunes of Thoratec, which has been a favorite of Wall Street investors, may depend on its ability to find an answer to the apparent jump in pump-related blood clots.

The HeartMate II has been a lifesaver for many patients in the final stages of heart failure, sustaining them until they get a heart transplant or permanently assisting their heart. Cheney received his device in 2010. Starling said that he planned to keep using the HeartMate II in appropriate patients because those facing death from heart failure had few options.

But the company has also been pushing to expand the device’s use beyond patients who face imminent death from heart failure. For example, the FDA approved a clinical trial for patients with significant, but less severe, heart failure to receive a HeartMate II to compare their outcomes with patients who take drugs for the same condition. Researchers at the University of Michigan Medical Center who are leading the trial said Wednesday that, based on the lower rates of blood clots seen in the Intermacs registry, they are planning to move forward with the trial.

Starling and researchers at Cleveland Clinic tried this spring to get The New England Journal of Medicine to publish a report about the findings at that hospital, but the publication declined, saying the data might simply represent the experience of one facility. As a result, Starling contacted Duke University and Washington University for their data. When analyzed, it mirrored events at Cleveland Clinic, he said.

The problems seen with the HeartMate II at the three hospitals were continuing as recently as this summer, when researchers paused the collection of data to prepare Wednesday’s study. The study also noted that a preliminary analysis of data provided by a fourth hospital, the University of Pennsylvania, showed the same pattern of blood clot formation but that the data had been submitted too late for full analysis.