VCU news

By Lorraine Cichowski and Michael Kontos

RICHMOND, Va. – In one of the largest and most-diverse studies of chest pain patients, researchers at the Virginia Commonwealth University Medical Center have shown that low levels of the cardiac marker, troponin I, are a poor predictor of a patient’s eventual outcome and result in frequent misdiagnoses that could delay treatment for coronary heart disease.

The study, conducted at the VCU Medical Center between June 1996 and March 2000, tracked 4,123 chest pain patients who underwent testing for their troponin I levels after being admitted to the coronary care unit for evaluation. Troponin is a protein found in the heart that is regarded by clinicians as one of the most-specific biochemical markers for heart damage.

The findings, reported in the March 17 issue of the Journal of the American College of Cardiology, indicated that patients with the highest troponin I values had the worst outcomes when assessed 30 days and again in six months after their initial evaluations. Of the 425 patients (10 percent of the total study group) with the highest troponin I values, 7 percent died of heart-related disease, 72 percent had a heart attack and 44 percent underwent revascularization treatment to restore blood flow to the heart. The findings are consistent with clinical trials of high-risk coronary patients that demonstrated the diagnostic value of troponin and led a joint U.S./European cardiology committee to recommend that patients with detectable troponin be diagnosed as having had a heart attack and treated accordingly.

The VCU study extends the findings to lower-risk patients, typical of those who traditionally come to hospital emergency departments for evaluation of chest pain as opposed to the higher-risk patients with diagnosed coronary disease usually enrolled in clinical trials. It found that, like the higher troponin I values, low troponin I values also predicted cardiac events, but at a much lower rate. The 270 patients (6.5 percent of the study total) with low troponin I values were three times more likely than patients with no detectable troponin I to die and more than five times as likely to be diagnosed as having a heart attack. However, these patients also were much more likely to have no significant coronary disease when angiography was performed and more likely to have a negative stress test when compared with those patients who had higher troponin I values.

“Cardiac troponin is the gold standard test for heart attacks. If a test finds that it’s significantly elevated, you’ve had a heart attack,” says Michael C. Kontos, M.D., assistant professor of internal medicine and lead author on the article.

“But the definition of ‘positive troponin levels’ varies among the different clinical trials. In addition, the trials have tended to enroll a majority of patients with a high prevalence of coronary disease, under representing women and the elderly who may not have a history of heart problems. The conclusions, therefore, have been difficult to apply, particularly when evaluating patients in the emergency department who may have lower troponin concentrations.” 

Kontos said the VCU study indicates that, in most cases, low troponin I values should not be considered diagnostic of a heart attack. However, given the higher adverse event rate compared with patients with no detectable levels of troponin, it would be prudent to consider further evaluation and tests in chest pain patients with low troponin I values to avoid missing a diagnosis of heart disease.

Parts of the study were presented at the American Heart Association Scientific Meetings in November 2002.