Blood test can predict long-term risk of death in patients with shortness of breath

Researchers from Massachusetts General Hospital (MGH) have shown that a blood test previously found useful for diagnosing heart failure and determining prognosis in several cardiovascular conditions can also predict the risk of death among patients coming to hospital emergency departments with shortness of breath. Their report in the Feb. 14 Archives of Internal Medicine describes how measuring levels of a protein called NT-proBNP was the most powerful predictor of whether patients would die within the year.

“We found that, regardless of whether the patient had been diagnosed with heart failure, elevated levels of this protein were strongly prognostic for death at one year and low levels identified those at the lowest risk,” says James Januzzi Jr., MD, of the MGH Cardiology Division, the paper’s lead author. “Based on these results, we recommend that every patient coming to a hospital with shortness of breath be tested for NT-proBNP.”

NT-proBNP is a natriuretic peptide, a protein produced when the cardiac muscle is under stress. In recent years these proteins have been investigated as assays for various forms of cardiovascular disease. In April 2005, members of the same MGH research team published the PRIDE study, which enrolled almost 600 patients who came to the MGH Emergency Department with shortness of breath. The PRIDE study showed that emergency department testing for NT-proBNP could help diagnose heart failure in those patients, providing a definitive diagnosis when the result was above specific levels and ruling it out when results were below an identified point.

The current study, designed to evaluate the long-term prognostic usefulness of the test, followed up on the patients in the PRIDE study. The research team contacted primary care physicians or cardiologists for each of the study participants, or reviewed hospital records, to determine their status a year after the original emergency visit. The results, which included data from all but four of the PRIDE study participants, showed that the NT-proBNP measurement made in the emergency department was a more accurate predictor of the risk of death during the following year than were other known measurements of cardiac risk.

“One of the more impressive findings in our study was the remarkably strong prognostic ability of NT-proBNP measurement in patients with shortness of breath from all causes,” Januzzi added. “An elevated NT-proBNP was as strong a predictor of death in patients without heart failure as it was for those with it. And remarkably, among those found to have heart failure in the original visit, no patient with an NT-proBNP level below a certain point died during the study period.

“Elevations of NT-proBNP can be used not only to determine the diagnosis and the best hospital setting for treatment but also to estimate longer-term prognosis, allowing us to identify patients needing more intensive treatment and follow-up,” he adds. “The next logical step is to evaluate treatment interventions for those with higher NT-proBNP levels to see if we can reduce their considerable risk for death.” Januzzi is currently directing a randomized clinical trial of medical therapy guided by NT-proBNP measurement for patients with recent heart failure.

The study’s coauthors are Rahul Sakhuja, MD, MPP, Michelle O’Donoghue, MD, Aaron Baggish, MD, Saif Anwaruddin, MD, Claudia Chae, MD, MPH, Renee Cameron, Daniel Krauser, MD, Roderick Tung, MD, and Carlos Camargo Jr., MD, of MGH, and Donald Lloyd-Jones, MD, ScM, of the Feinberg School of Medicine at Northwestern University. The study was supported by a grant from Roche Diagnostics, which manufacturers the NT-proBNP assay studied but had no role in conducting the investigation or reporting the results.