Author: Jon Merz
Date: 08-05-04 10:22
Source: Wall St J
URL: http://online.wsj.com/article_print/0,,SB109164588494382980,00.html
Date published: August 5th 2004
From Haavi Morreim:
Heralded Heart Drug
Has Unexpected Risks
Hospitalizations, Deaths Highlight Hazards
Of Applying Study Results to Individual Patients
By RON WINSLOW
Staff Reporter of THE WALL STREET JOURNAL
August 5, 2004; Page D1
A drug that was shown in a landmark clinical trial five years ago to save
lives of patients with heart failure may have turned out in actual medical
practice to have taken lives instead.
The unexpected turn of events, reported in today's New England Journal of
Medicine, doesn't raise doubts about the drug's efficacy, but offers a
cautionary tale for doctors and patients alike as they seek to apply the
findings of clinical drug studies to real-world medicine.
In 1999, researchers astonished the medical world with a study showing that
a little-used, 40-year-old drug significantly reduced death and
hospitalization for patients with congestive heart failure. Those results,
which indicated the drug prevented or postponed 11 deaths and eight hospital
admissions for every 100 patients treated, were so compelling the trial was
halted ahead of schedule and the New England Journal of Medicine released
the results online, weeks before its actual publication.
Now, Canadian researchers are reporting that the drug, called
spironolactone, may have had a much different effect. Publication of the
1999 study resulted in a more than fourfold jump over an 18-month period in
prescriptions for the generic drug, according to the researchers. That surge
in use was accompanied by a tripling of hospital admissions and of deaths
resulting from dangerous elevations of potassium -- a known side effect of
spironolactone.
POOR OUTCOMES
Among heart patients studied by Canadian researchers:
• Prescriptions for spironolactone rose to 149 per 1,000 patients in 2001,
from 34 per 1,000 before 1999.
• Hospital admissions for hyperkalemia rose to 11.0 per thousand in 2001,
from 4.0 per 1,000 patients in early 1999.
• Deaths attributed to hyperkalemia rose to 2.0 per thousand in 2001, from
0.3 per thousand in 1994.
Source: The New England Journal of Medicine
In the 1999 study, the incidence of this side effect, called hyperkalemia,
was 2% and not statistically different from patients who got a placebo. The
surge of use following the study was associated with at least 560 additional
hospital admissions and 73 additional deaths from hyperkalemia, by one
measure, the Canadian researchers found.
Researchers said the main culprit wasn't the drug itself, but rather the
willingness of doctors to prescribe it for patients who, for a variety of
reasons, already were at heightened risk for the potassium problem, and then
weren't carefully monitored for the side effect. The drug, marketed by Mylan
Laboratories Inc., of Canonsburg, Pa., had been mainly used as a diuretic
and for liver patients before the 1999 study.
"We have to blame the overly enthusiastic embrace of the study's findings
for what we're seeing now," says David Juurlink, a scientist at the
Institute for Clinical Evaluative Sciences, Toronto, and the University of
Toronto. Many patients given the medicine likely would have been excluded
from participating in the clinical trial, Dr. Juurlink says.
The new findings offer a provocative look at the difference between clinical
trials and real-world medicine -- and the potential dangers of applying
trial results too widely. Patients in clinical studies typically are
carefully selected to maximize the chance of showing a benefit and minimize
side effects. Thus trial patients represent only a subset of the types of
patients doctors treat in their offices.
"Everyone assumes that what you get in a clinical trial is what you will see
in clinical practice," says Milton Packer, a clinical-trial and
heart-failure expert at University of Texas Southwestern Medical Center,
Dallas. But "the broader population is always at greater risk for side
effects than patients in a clinical trial."
For instance, Rezulin, a diabetes drug once marketed by the former Warner
Lambert Co., was shown to be safe in clinical trials, but once it was
prescribed to patients in the broader population, it was linked to rare but
serious liver side effects that led eventually to its being withdrawn from
the market.
In the new study on spironolactone, researchers analyzed prescription-drug
and hospital-admissions databases covering more than 1.3 million Canadians
age 66 or older. They found that patients given the medicine in the
aftermath of the 1999 study were on average 13 years older than participants
in the original trial and more likely to have diabetes. Older patients and
diabetics are at higher risk for elevated potassium. Also, the average dose
in actual practice was 30 milligrams, while 25 milligrams was used in the
study, Dr. Juurlink says, perhaps reflecting how doctors turn to higher
doses if an initial lower dose is ineffective.
Highly elevated potassium can cause severe muscle pain. Dr. Juurlink says
one recent patient who was driven to the hospital was unable to get out of
the car on his own. More troubling, abnormally high potassium can disrupt
electrical conduction in the heart, causing sudden death.
Pfizer Inc. recently began marketing eplerenone, a second-generation version
of spironolactone, for patients with heart failure after a heart attack.
Officials said marketing materials include detailed advice to doctors on how
to choose patients and monitor potassium levels to minimize occurrence of
any adverse effects.
The new report underscores the increasing complexity in the treatment of
heart-failure patients, who are typically on multiple medications and suffer
from additional conditions -- all of which can increase potassium levels or
hinder its clearance from the body. That makes applying evidence from
clinical trials even more difficult.
For instance, drugs called ACE-inhibitors, similar drugs called angiotensin
receptor blockers, or ARBs, and beta blockers all have grown in use for
heart failure since 1999, and all can contribute to a rise in blood levels
of potassium. Also, says Biff F. Palmer, an internist at University of Texas
Southwestern Medical Center, as many as one-half of heart-failure patients
have impaired kidney function, which can restrict their ability to clear
potassium.
Doctors considering adding spironolactone to a patient's regimen had all of
these issues to consider, and the 1999 study offered little guidance on how
other drugs or health conditions factored in.
In Toronto, Dr. Juurlink and his colleagues sought to adjust for these other
factors when combing through the Canadian patient databases. The study isn't
ironclad proof of cause and effect, in part because researchers didn't
specifically match individual patients taking the drug with poor outcomes.
Dr. Juurlink says there isn't any other obvious reason to explain the
findings.
He also notes that while the results are based on patients in Ontario, he
believes the 1999 study sparked similar increases in use of spironolactone
in the U.S. If he is right and doctors similarly went beyond the parameters
of the study in their prescribing, the Canadian findings would correspond in
the U.S. to 37,000 additional hospital admissions and 4,200 additional
deaths each year from potassium-related causes.
Dr. Juurlink stressed that he is convinced that appropriate patients reap
important benefits from spironolactone and that the risk can be managed by
carefully selecting patients, monitoring their potassium and adjusting
medications to minimize adverse effects. "This is not an indictment of the
study," he says. "It is how it was translated into clinical practice."
Write to Ron Winslow at ron.winslow@wsj.com1
Copyright 2004 Dow Jones & Company, Inc. All Rights Reserved
|
|
