B-type natriuretic peptide is an accurate predictor of heart failure in the emergency department - Patient-Oriented Evidence that Matters - Brief Article

Journal of Family Practice, Oct, 2002 by John O'Connor, Linda N. Meurer

Maisel AS, Krishnaswamy P, Nowak RM, et al. Rapid measurement of B-type natriuretic peptide in the emergency diagnosis of heart failure. N Engl J Med 2002; 347:161-7.

* BACKGROUND B-type natriuretic peptide is released from the cardiac ventricles in response to volume expansion and pressure overload. Levels correlate with severity of congestive heart failure (CHF) and prognosis. A rapid assay for B-type natriuretic peptide might help clinicians' accuracy in distinguishing CHF from other conditions (eg, chronic obstructive pulmonary disease) as the cause of acute dyspnea in an emergency setting.

* POPULATION STUDIED The investigators enrolled 1586 patients from 7 sites (5 in the United States, 1 in France, 1 in Norway). Eligible subjects (at least 18 years old) presented to the emergency department with shortness of breath for which CHF court not be obviously ruled out (such as in trauma or cardiac tamponade). Patients were not included if they had acute myocardial infarction or renal failure; patients were also not included if they had unstable angina, unless their predominant presenting symptom was dyspnea. Forty-four percent of the subjects were women; 49% were white, 45% were black, and 6% were from other races.

* STUDY DESIGN AND VALIDITY This study evaluated the role B-type natriuretic peptide determinations might play in the diagnosis of CHF by comparing B-type natriuretic peptide levels with clinical diagnosis as the gold standard. Blood was drawn from all included patients for measurement of B-type natriuretic peptide. Emergency room physicians, who were not given the laboratory results, assessed the probability that the patient had CHF. Patients with a history of CHF were classified as having either an exacerbation of CHF or dyspnea from another cause with underlying left ventricular dysfunction.

To determine the actual diagnosis, 2 cardiologists independently reviewed all medical records pertaining to the patient and classified the diagnosis as dyspnea due to CHF, acute dyspnea due to noncardiac causes in a patient with a history of left ventricular dysfunction, or dyspnea not due to CHF. The cardiologists were blinded to the B-type natriuretic peptide level as well as to the emergency department physicians' diagnoses. The cardiologists also had access to the radiologists' report on the chest x-ray; past medical history from old charts; and subsequent test results such as echocardiography, radionuclide angiography, and left ventricular angiography if performed, and the hospital course for patients admitted to the hospital.

This study was well designed. The population was inclusive and appropriate to practice, although it is unclear if all patients who might be have been eligible were enrolled. The test, a rapid assay of sampled blood, was reasonable and acceptable to patients. The results of the B-type natriuretic peptide blood levels were not known to either the emergency department physicians who made the initial diagnosis of CHF or to the cardiologists who confirmed the diagnosis. Although the study measured the immediate diagnostic value of the test, it did not evaluate patient-oriented outcomes such as cost, or whether the test had an effect on treatment decisions, longevity, or quality of life.

* OUTCOMES MEASURED Whole blood or plasma levels of B-type natriuretic peptide were measured ruing a fluorescence immunoassay kit (Triage BNP Test; Biosite Inc, San Diego, CA) and the results were compared with clinical diagnoses to determine the sensitivity, specificity, and accuracy of the test in the diagnosis of CHF. Receiver-operating-characteristic curves were constructed to illustrate various cutoff values of B-type natriuretic peptide. Long-term outcomes of patients with CHF were not measured.

* RESULTS Congestive heart failure was diagnosed in 744 patients (47%), dyspnea due to noncardiac causes in 72 patients with a history of CHF (5%), and no CHF in 770 (48%). The B-type natriuretic peptide level test performed well for diagnosing CHF; the area under the receiver-operating-characteristic curve was 0.91 (where 1.0 indicates a perfect test). A value of 100 pg/mL or more was the single most accurate predictor of the presence of CHF when compared with clinical predictors such as history, physical examination, or chest x-ray. This B-type natriuretic peptide cutoff value of 100 pg/mL was 90% sensitive, 76% specific, and 83% accurate in differentiating CHF from other causes of dyspnea. As such, this cutoff value outperformed the accuracy of 2 commonly used clinical criteria used for diagnosing CHF, the National Health and Nutrition Examination Survey (NHANES) criteria (67%) and Framingham criteria (73%). At this prevalence of 47%, a level of 50 pg/mL was associated with a negative predictive value of 96%. B-type natriuretic peptide values also correlated with CHF severity as determined by the New York Heart Association functional class.

RECOMMENDATIONS FOR CLINICAL PRACTICE

The B-type natriuretic peptide level assay is a rapid (15-minute) whole blood test that can be done at the bedside or in the emergency department to diagnose congestive heart failure as the cause of acute dyspnea. Using a cutoff of 100 pg/mL, the test has better accuracy than either NHANES criteria or Framingham criteria. However, whether use of this new test will improve patient outcomes is unknown.