Blood Test Helps Identify Heart Patients At Risk Of Dying
DURHAM, N.C. -- A blood test that measures protein expelled
from damaged heart tissue is the best indicator yet of which
patients are at risk of dying within weeks of a suspected heart
attack, an international research team led by Duke University
Medical Center reported Wednesday.
The study, conducted in 96 United States and Canadian
hospitals and published in the Oct. 31 issue of the New England
Journal of Medicine, shows that the "troponin T" test is a
powerful indicator of how injured a heart is, especially when
used in patients who don't show clear evidence of trouble from
standard electrocardiography (EKG) readings.
The researchers say the study is important because
physicians can't easily detect distinct signs of heart disease
in some patients who come to the hospital with chest pain. The
American Heart Association estimates that in one year as many
as 1.5 million Americans will have a heart attack and 500,000
of them will die. Thousands of those who will die will have
been prematurely released from the hospital because their heart
attack was not diagnosed, the study's authors said.
Troponin T is a regulatory protein that helps heart muscle
contract. Once this muscle starts dying, cell membranes break
down and internal cellular material, including proteins, are
released and are carried away by blood.
Authors of the study say that the troponin T test offers a
single measurement that can help emergency department
physicians and cardiologists more readily determine which heart
patients should be referred for more comprehensive diagnostic
"The relationship between the blood level of this protein
and the risk of death was a direct one in this study," said Dr.
Magnus Ohman, the lead researcher and a cardiologist at the
Duke Clinical Research Institute. "The troponin T test is an
extremely useful tool that can give physicians additional
insight into the extent of a patient's heart disease.
"In the least, it should be used in a hospital's emergency
department when other indicators, such as an electrocardiogram,
don't show clear evidence of trouble," Ohman said in an
Co-author Dr. Robert Califf, director of the Duke institute,
said that while the test may result in increased health care
costs, because it will identify those patients who need
immediate and aggressive heart care, it "will save more
"We have long needed a way to identify those patients with
no telltale signs of possible acute heart disease other than
pain," Califf said. "In an emergency department, we need
additional tools to help us decide who should receive quicker
The test was administered to 855 patients in North America,
and involved physicians from both small community hospitals and
large research centers. The primary researchers were from Duke,
The Cleveland Clinic Foundation, The University Of Edmonton in
Canada, the University of Maryland Medical System, the
Universities of Heidelberg and Hamburg in Germany, and
Methodist Hospital-Baylor College of Medicine in Houston.
It was conducted as a sub-study of a larger clinical trial
known as GUSTO IIa, and was supported by Ciba-Geigy Corp.,
Advanced Cardiovascular Systems, and the German pharmaceutical
company Boehringer Mannheim, which manufactures the troponin T
test. The test, which costs $30, received U.S. Food and Drug
Administration approval in 1994.
In the study, the troponin T test was administered within 12
hours of the onset of symptoms in patients, but physicians did
not use test results to refine treatment. Rather, the tests
were "blinded," that is, they were processed at a lab at the
University of Maryland Medical System, and results were not
known until all patients in the study had been enrolled.
These results were then compared to indications of heart
disease from an EKG, which records the flow of electrical
energy through the heart, and from a standard blood test known
as CK-MB, which measures creatine kinase, another marker of
heart damage. CK-MB must be administered at least twice, once
when a patient is first seen, and then several hours later for
a comparison reading. All three test measurements were then
compared to what treatment the patient received and whether
Specifically, the researchers found:
· On admission to the hospital, 36 percent of the patients
had elevated troponin T levels. Death within a month was
significantly higher in these patients (11.8 percent) than in
patients with lower levels of the protein (3.9 percent).
· Of the three measurements, troponin T level was the
variable most strongly related to 30- day mortality, followed
by EKG readings, and then CK-MB test results. The relation
between protein level and death rate was "nearly linear and
direct," Ohman said.
· Patients with a positive troponin T level but with no
significant EKG readings were more than three times as likely
to die in a month than those patients with low troponin T and
negative EKG readings. Specifically, among the patients with no
clear evidence of new heart damage, 9.9 percent of those with
positive troponin T results died compared to 3 percent of
patients with low troponin T.