Friday May 16, 2003

By Bruce Taylor Seeman
Newhouse News Service

WASHINGTON -- A dispute among prominent researchers has mushroomed into a rancorous national debate about how patients ought to be protected in medical experiments.

Central to the conflict is a compelling question: Did design flaws in a landmark series of clinical trials put some patients at greater risk of death?

The nation's medical researchers may be forced to rethink basic rules for studies involving vulnerable patients.

The dispute stems from a sweeping investigation of a respected research team that has spent more than $37 million in federal money to test therapies for acute respiratory distress syndrome, or ARDS, an often fatal inflammatory lung condition.

The federal government's Office for Human Research Protections last summer forced the shutdown of one study by the team, which is known as the ARDS Network or ARDSNet.

Since then, the dispute has been a hot topic in medical journals, symposia and the National Institutes of Health, the nation's foremost center for clinical research.

In Louisiana, Charity, Ochsner and Tulane University hospitals participated in the study. It was also conducted at 19 other medical centers around the country.

Bennett deBoisblanc, a doctor at Charity Hospital who oversaw the study, said he disputes the information that caused it to be put on hold, but will comply until he is allowed to move forward. In the meantime, he said, 400 patients have enrolled in the study and "we're hoping it will start back up and we'll be able to complete the trial."

The federal probe began when scientists at the National Institutes of Health charged that the ARDSNet team made flawed use of a basic scientific tool, the control group.

In a classic medical experiment, some patients receive the therapy being tested while others, in the control group, get the current standard treatment. The control group allows researchers to make meaningful comparisons.

But Drs. Charles Natanson and Peter Eichacker, the NIH scientists who have led the criticism of ARDSNet, charge in a letter in the March 1 edition of the American Journal of Respiratory and Critical Care Medicine that during one study of 861 patients, control group subjects died at a significantly higher rate -- 40 percent -- than eligible patients who refused to participate, among whom the death rate was 31 percent. ARDSNet's leaders declined to discuss the letter.