Studies Bolster Link Between Diet Drugs, Heart-Valve Leaks

By Laura Johannes and Robert Langreth
Copyright 1998 Wall Street Journal
September 10, 1998

Three studies in Thursday's New England Journal of Medicine, including one funded by a company that sold diet pills, provide the strongest evidence yet linking long-term use of the diet drugs Redux and fenfluramine to heart-valve leaks.

But all three find that the problems affect significantly fewer than the 30% of users estimated by the Food and Drug Administration when it asked American Home Products Corp. to withdraw the drugs a year ago.

Most surprising is the formal publication of a study funded by American Home, which seemed to find little evidence of heart-valve problems in short-term Redux users when it was first presented last March. Now the same data are presented differently, and the study's author agrees the results show possible hints of early valve disease, especially in the context of the other studies.

The studies are important because they are among the first linking the diet drugs to heart problems to be reported in a top scientific journal since the recall. "All three studies are concordant and document that [valve disease in diet-drug users] is very real," said Ann Bolger, a cardiologist and spokeswoman for the American Heart Association. "This could be an epidemic we are dealing with over a long period of time."

Long-Term Significance

Still, scientists said the long-term medical significance of the heart-valve problems is unclear, especially since very few cases have required surgery. "If there is anything there at all, it is pretty minuscule," said Nelson Schiller, a heart expert at the University of California at San Francisco. The diet drugs, he added, "might be an acceptable risk for some people."

American Home officials didn't dispute the results, but noted that the clinical significance of the valve leaks was unknown, as most of those affected have no symptoms and researchers detected them only through ultrasound tests. "There were few reports of clinical signs or symptoms related to heart-valve abnormalities," said Dr. Philip de Vane, vice president for clinical affairs at American Home's Wyeth-Ayerst Laboratories unit. American Home, Madison, N.J., is expected to finish up a large study of patients who used fenfluramine and another diet drug, phentermine, that could shed light on the issue by year end.

The publication in the prestigious medical journal could bolster numerous lawsuits against American Home. "It unquestionably will help us that a company-funded study found measurable injury from the drugs," said Alex MacDonald, an attorney at the Boston law firm of Robinson & Cole.

Firm Calls Study 'Moot'

Interneuron Pharmaceuticals Inc., the Lexington, Mass., company that developed Redux and contracted with American Home to market it, said the study results are "moot" for it since has negotiated a $100 million settlement agreement that, if approved by courts, would absolve it of all further liability.

When initially presented at a March press conference, the American Home-funded study was viewed as vindication of Redux. Author Neil J. Weissman of the Georgetown University Medical Center said at the time that he found "no difference" in significant valve disease between patients who took the drug for two or three months and those taking a sugar pill, and called the results "reassuring for the majority of patients who have been on Redux."

But the same data in the New England Journal of Medicine article are presented very differently, because of changes demanded by its editors as a condition for publication. The published paper emphasizes that, when all types of heart-valve leaks were considered, even ones too small to be normally considered significant, the diet-drug patients did fare worse than sugar-pill takers. Seventeen percent of Redux takers had aortic valve leaks, compared with 12% of those taking the placebo, and 61% had mitral valve leaks, compared with 54% of those on the placebo.

Combination of Reports

Dr. Weissman said he initially stressed the fact that the drug didn't cause any significant leaks because that was most newsworthy at the time, given government estimates that the problems occurred in more than 30% of diet-drug patients. Now, in seeing the two other papers, he said he understands why the journal insisted on emphasizing his conclusion that Redux was linked to smaller leaks. "Putting [my data] with other papers implies it may be the beginning of something," he said. But he added that it is impossible to know for sure whether the small leaks would intensify without giving test subjects the recalled drugs, which would be unethical.

In the second study, first presented in preliminary form late last year, Mehmood Khan and colleagues at Hennepin County Medical Center in Minneapolis performed heart ultrasounds, or echocardiograms, on 233 patients who took Redux or fenfluramine, often in combination with phentermine, and compared the results to a control group of obese patients who hadn't taken diet drugs. They found that 23% of the drug patients had heart-valve abnormalities, compared with 1.3% of the control group. This is somewhat less than the 32% rate of heart problems that the FDA found at the time of the recall, but still high.

A third study, an analysis of medical records of about 10,000 diet-drug takers in England, found 11 cases of newly diagnosed symptomatic heart-valve leaks, compared with none for a similar group that didn't take the drugs. That study showed a much lower prevalence of leaks than the Khan study because it counted only those patients who had symptoms, said author Hershel Jick, an associate professor of medicine at Boston University Medical School. Symptoms ranged from fatigue to chest pain and heart murmurs.

Comments on this posting?

Click here to post a public comment on the Trash Talk Bulletin Board.

Click here to send a private comment to the Junkman.

Material presented on this home page constitutes opinion of Steven J. Milloy.
Copyright © 1998 Steven J. Milloy. All rights reserved on original material. Material copyrighted by others is used either with permission or under a claim of "fair use." Site developed and hosted by WestLake Solutions, Inc.