BNP: A Hearty Diagnostic Complement to the Chest Film in Heart Failure

Abstract & Commentary

Source: Knudsen CW, et al. Diagnostic value of B-type natriuretic peptide and chest radiographic findings in patients with acute dyspnea. Am J Med 2004;16:363-368.

Recently, enthusiasm has emerged for a new investigative assay for beta-type natriuretic peptide (BNP), a cardiac neurohormone released by atrial and ventricular myocytes undergoing volume overload or stretch. BNP has been demonstrated to be of value in the diagnosis of congestive heart failure (CHF); it rises to higher levels in more severe CHF conditions and correlates with pulmonary capillary wedge pressure. To evaluate the complementary roles of BNP and chest x-ray (CXR) in the evaluation of CHF, Knudsen and colleagues studied an international cohort of emergency department (ED) patients with acute dyspnea.

A total of 880 patients was analyzed as a subset of the heralded Breathing Not Properly Multinational Study, conducted at five U.S. teaching hospitals and two academic centers in Paris and Oslo in 1999-2000. Patients were seen in these EDs with acute or suddenly worsening dyspnea. The average age was 64; 53% were male; and 56% were black. Those with acute myocardial infarction, renal failure, trauma, or pneumothorax were excluded. ED physicians performed clinical assessment and electrocardiogram interpretation. Radiologists read all CXRs. BNP levels were determined by point-of-care testing in triplicate, but investigators were blinded as to the results.

Thirty days later, the clinical, laboratory, and radiologic data for each patient were reviewed by two independent cardiologists using the Framingham scoring criteria to adjudicate each case as either dyspnea due to CHF or non-cardiac dyspnea. Based on these analyses, the most frequent final diagnosis was acute CHF in 447/880 cases (51%). Mean BNP levels were significantly higher in CHF patients than in non-cardiac dyspnea cases (689 pg/mL, vs 121 pg/mL; p <0.0001). In acute CHF patients, 401/447 (90%) had BNP levels greater than 100 pg/mL, vs 108/433 cases (25%)  without CHF. Presence of cardiomegaly, cephalization, interstitial edema, and alveolar edema on CXR had specificities for CHF of 80, 96, 98, and 99%, respectively. Logistic regression analysis demonstrated that BNP levels greater than 100 pg/ml enhanced the diagnostic accuracy for CHF beyond the CXR findings alone (OR 21.4, 95% CI 14.6-31.3). Further analysis revealed that a BNP level exceeding 100 pg/ml, (OR 12.3), coupled with CXR evidence of cardiomegaly (OR 2.3), cephalization (OR 6.4), and interstitial edema (OR 7.0), greatly enhanced the accuracy of CHF diagnosis beyond that attained by clinical assessment alone (p < 0.001 for all ORs). The authors conclude that BNP measurements are complementary to CXR in the detection of CHF in acutely dyspneic patients.

Commentary by Michael Felz, MD

BNP levels generated helpful diagnostic information not available from clinical assessment alone. In fact, BNP level was the single strongest statistical predictor of acute CHF, prompting the authors to suggest that BNP assays should become routine measurements in ED patients with acute shortness of breath.

I find this data quite encouraging. It now is possible, and may be wise, to determine BNP values at the very outset of evaluation of the dyspneic ED patient, even while awaiting results from the usual clinical, laboratory, and radiologic investigations. The predictive value of elevated BNP levels greater than 100 pg/mL, and certainly higher values—such as the average of 689 pg/ml demonstrated in the 447 patients proven to have CHF—would provide persuasive, early diagnostic data that could allow therapeutic interventions (and blessed relief from air hunger) to happen more quickly and efficiently, in the ED. Conversely, a low BNP level (less than 100 pg/ml) would steer me toward non-CHF causes of dyspnea, such as asthma or pulmonary embolus. My impression is that BNP is here to stay as a hearty complement to CXR in the diagnosis of CHF.

Dr. Felz, Associate Professor, Department of Family Medicine, Medical College of Georgia, Augusta, is on the editorial board of Emergency Medicine Alert.