Drug's clinical trial success paved way to overprescribing

Five years after the publication of a large trial that added a medication to the drug regimen of many heart failure patients, physicians are discovering that a good study doesn't necessarily translate into improved outcomes.

"The Randomized Aldactone Evaluation Study (RALES) was very impressive and made a major change in how we were treating heart failure," said Stephen Siegel, MD, a cardiologist and clinical assistant professor of medicine at New York University School of Medicine. "But there are differences between management in an academic, controlled environment and the usual clinical setting."

According to a paper published last month in the New England Journal of Medicine, the NEJM's 1999 publication of RALES did lead to an increase in prescribing of the drug but did not result in the study's 30% decrease in mortality rates. Hospitalization for hyperkalemia, a known side effect of the drug, increased, as did the deaths from this complication. This increase was greater than expected from the uptick in the drug's use.

The paper's authors say that the original trial was excellent and justified changing clinical practice. Still, the chain of events was far from perfect.

"This is a life-saving drug for patients with bad heart failure, and the RALES trial showed us that," said David Juurlink, MD, PhD, lead author of the latest paper and a scientist at the Institute for Clinical Evaluative Sciences in Toronto. "But a couple things went wrong."

Dr. Juurlink suggested that the same dramatic results were not seen in the real-world setting because doctors were less likely to check for side effects than they were in the clinical trial. Also, in actual practice, the drug was prescribed at higher doses.

In addition, the real-world patients who received the drug were in many ways different from those included in the study. In the original trial, patients had severe heart failure. But after the trial's publication, physicians started prescribing the drug to more mild cases or to those who had different kinds of heart problems. The drug was prescribed to those who were older, already had mild hyperkalemia or had some impairment of kidney functioning. It was also prescribed to patients who were on potassium supplementation or other drugs that are linked to hyperkalemia.

"We extended the findings of the trial a bit wider than we should have," Dr. Juurlink said. "And we didn't monitor patients as closely as we should have."

Experts say, however, that it's not the clinical trial setting that needs to change. The real world does.

"Physicians in the community embraced the findings but applied them to patient types that would never have been allowed to be entered in the RALES trial," said Biff F. Palmer, MD, a nephrologist and professor of internal medicine at University of Texas Southwestern Medical School. He wrote a review article in the same issue of the NEJM about dealing with hyperkalemia. "But physicians also don't have the time in a clinical environment to counsel patients not to take certain medicines. You don't have the luxury of frequent follow-ups, and these are all the things that are required when you're treating a higher-risk patient population."

Lost in translation

To a certain extent, this situation does have unique qualities. Experts speculate that the transition of this new information into practice was particularly rough because spironolactone is a decades-old, cheap drug available in generic form. Consumers can buy a three-month supply for less than $20. Doctors were comfortable using it because it had such a long history, but, as a result, observers say it was not likely to be a topic of continuing medical education or drug representatives' physician education regarding its new uses.

"People were very, very familiar with it, and they may have become a little more cavalier about it," said Dr. Siegel. "And when you're dealing with a new use for a generic drug, you don't have the same force of pharmaceutical representatives."

Those who led the original trial caution that Dr. Juurlink's study should not restrict use of the drug by those who will benefit from it, but suggest that the missing piece was support for physician education.

"When used properly, this is a very good drug that saves people's lives and keeps people out of the hospital," said Bertram Pitt, MD, the lead author of the original 1999 paper and professor of medicine at the University of Michigan School of Medicine. "But all that postgraduate education that normally would occur with a new concept and new drug, never happened here. When a drug is generic, then there should be some funds from the NIH or somewhere else to provide some of the postgraduate education."

Meanwhile, this case illustrates just how tricky translating science into medicine can be and how important it is to improve that process.

Many medical societies and scientific institutions have announced programs to improve the translation of science into practice. Most recently, the AMA pledged at its December 2003 meeting to work to improve the translation of bench science to clinical practice and acknowledged at its 2004 Annual Meeting that the increasing pace of basic science was creating a bottleneck in the process.

"One wants to hope that the translation goes beyond just simply pointing out that a new drug or a new treatment in patients with disease X is better or worse," said John F. Schneider, MD, MPH, vice chair of the AMA's Council on Scientific Affairs.

"This clearly points out a role for effective continuing medical education to serve as a mechanism to translate the details of clinical research activities that have a positive impact on improving patient care."