RSS Feeds
Does B-Type Natriuretic Peptide Testing Affect Outcome and Management of Patients With Acute Dyspnea?
- Hans-Gerhard Schneider, MBBS, MD;
- Louisa Lam, MPH; and
- Peter Cameron, MBBS, MD
- From Alfred Pathology Service, Alfred Health, Prahran, Victoria 3181, Australia, and Monash University, Alfred Hospital, Melbourne, Victoria 3004, Australia.
IN RESPONSE:
We appreciate the responses to our article. Dr. Tapolyai asks whether the emergency department physicians ignored the BNP results, which provide “clear, objective, laboratory-proven evidence.” Levels of BNP are elevated in a range of conditions other than heart failure, such as pulmonary embolism, acute coronary syndromes, and acute myocardial infarction, and increase with atrial fibrillation, radiographic cardiomegaly, decreased blood hemoglobin concentration, decreased body mass index, and increasing age (1). Our population was older than the populations in some other studies. The diagnostic accuracy of our physicians was 80.5% overall. In the Breathing Not Properly study (2), clinicians had a diagnostic accuracy of 74% in patients with a high likelihood of heart failure. In that study, addition of BNP test results in a mathematical model increased diagnostic accuracy to 81%. At the optimal BNP cutoff value, the diagnostic accuracy for BNP in our study was 82% (Lokuge A, Lam L, Cameron P, Krum H, de Villiers S, Bystrzycki A, et al. BNP testing and the accuracy of heart failure diagnosis in the emergency department. In preparation.).
Dr. Pachika and colleagues wonder how the emergency department physicians incorporated the BNP value into their assessment and whether the initial and final diagnoses were different. The main use of the BNP value is to exclude heart failure as the cause of shortness of breath (negative predictive value of 90% at 100 pg/mL) (2). In our study, the accuracy of the BNP value at 101 pg/mL was 71% (92% sensitivity and 51% specificity). In the BNP-tested group, 89 patients had BNP values less than 100 pg/mL. Of these, 74 were clinically judged not to have heart failure as the cause of shortness of breath, whereas 15 were judged to have heart failure. After laboratory and other results were incorporated, this number decreased to 10 patients, a relatively small change.
Dr. Dipaola and colleagues ask whether the management of patients in the BNP-tested group who were ultimately identified as having heart failure was better than that of those in the control group. We performed subgroup analysis and found no differences apart from use of noninvasive ventilation, which is of benefit both in heart failure and in chronic obstructive pulmonary disease (Table). Therefore, knowledge of BNP did not markedly improve patient management. Remember that the studies looking at combining BNP testing and clinical assessment for heart failure diagnosis are usually post hoc analyses of results that are mathematically modeled (2, 3) and are not randomized, controlled trials like this study. We will perform further modeling to identify intermediate-risk patients, who might benefit from BNP measurement, and perhaps age-stratified BNP values might be helpful for further improving diagnostic performance of the BNP assay. In summary, knowledge of BNP value in our study did not improve management or length of stay, perhaps partly because of the high level of diagnostic accuracy of the clinicians.
Hans-Gerhard Schneider, MBBS, MD
Alfred Pathology Service, Alfred Health
Prahran, Victoria 3181, Australia
Louisa Lam, MPH
Peter Cameron, MBBS, MD
Monash University, Alfred Hospital
Melbourne, Victoria 3004, Australia
Article and Author Information
-
Potential Financial Conflicts of Interest: None disclosed.
