Author + information
- Received June 1, 2002
- Revision received October 21, 2002
- Accepted November 19, 2002
- Published online June 4, 2003.
- Alan S Maisel, MD, FACC* (, )
- James McCord, MD,
- Richard M Nowak, MD,
- Judd E Hollander, MD,
- Alan H.B Wu, PhD, MD, MBA,
- Philippe Duc, MD,
- Torbjørn Omland, MD, PhD,
- Alan B Storrow, MD,
- Padma Krishnaswamy, MD,
- William T Abraham, MD, FACC,
- Paul Clopton, MS,
- Gabriel Steg, MD,
- Marie Claude Aumont, MD,
- Arne Westheim, MD, PhD, MPH,
- Cathrine Wold Knudsen, MD,
- Alberto Perez, MD,
- Richard Kamin, MD,
- Radmila Kazanegra, MD,
- Howard C Herrmann, MD, FACC,
- Peter A McCullough, MD, MPH, FACC,
- Breathing Not Properly Multinational Study Investigators*
- ↵*Reprint requests and correspondence:
Dr. Alan S. Maisel, VAMC Cardiology 111-A, 3350 La Jolla Village Drive, San Diego, California 92161, USA.
Objectives This study examines B-type natriuretic peptide (BNP) levels in patients with systolic versus non-systolic dysfunction presenting with shortness of breath.
Background Preserved systolic function is increasingly common in patients presenting with symptoms of congestive heart failure (CHF) but is still difficult to diagnose.
Methods The Breathing Not Properly Multinational Study was a seven-center, prospective study of 1,586 patients who presented with acute dyspnea and had BNP measured upon arrival. A subset of 452 patients with a final adjudicated diagnosis of CHF who underwent echocardiography within 30 days of their visit to the emergency department (ED) were evaluated. An ejection fraction of greater than 45% was defined as non-systolic CHF.
Results Of the 452 patients with a final diagnosis of CHF, 165 (36.5%) had preserved left ventricular function on echocardiography, whereas 287 (63.5%) had systolic dysfunction. Patients with non-systolic heart failure (NS-CHF) had significantly lower BNP levels than those with systolic heart failure (S-CHF) (413 pg/ml vs. 821 pg/ml, p < 0.001). As the severity of heart failure worsened by New York Heart Association class, the percentage of S-CHF increased, whereas the percentage of NS-CHF decreased. When patients with NS-CHF were compared with patients without CHF (n = 770), a BNP value of 100 pg/ml had a sensitivity of 86%, a negative predictive value of 96%, and an accuracy of 75% for detecting abnormal diastolic dysfunction. Using Logistic regression to differentiate S-CHF from NS-CHF, BNP entered first as the strongest predictor followed by oxygen saturation, history of myocardial infarction, and heart rate.
Conclusions We conclude that NS-CHF is common in the setting of the ED and that differentiating NS-CHF from S-CHF is difficult in this setting using traditional parameters. Whereas BNP add modest discriminatory value in differentiating NS-CHF from S-CHF, its major role is still the separation of patients with CHF from those without CHF.
As many as 40% to 55% of patients with signs and symptoms of congestive heart failure (CHF) have preserved systolic function (1,2). Cardiac abnormalities in these patients are determined by a complex sequence of interrelated events that may make diagnosis and the success of treatment difficult to assess (3). While Doppler echocardiography has been used to examine left ventricular (LV) filling dynamics in these patients, the limitations of this technique suggest the need for other objective measures in CHF patients with preserved systolic function (4).
B-natriuretic peptide (BNP) is a cardiac neurohormone secreted from the ventricles in response to ventricular volume expansion and pressure overload (5,6). B-type natriuretic peptide levels are known to be elevated in patients with symptomatic LV dysfunction and correlate to New York Heart Association (NYHA) class as well as prognosis (7–12). Recently, it has been shown that in patients with preserved LV function, BNP levels may be reflective of diastolic filling abnormalities on echocardiography (13,14).
This study examines BNP levels in patients presenting to the emergency room with shortness of breath as part of the multinational Breathing Not Properly study (15). To define and differentiate characteristics of those with non-systolic dysfunction from those patients with systolic dysfunction, we utilized that subset of patients who had echocardiographic determination of cardiac function within 30 days of their initial visit.
The study was approved by the Institutional Review Boards of participating Breathing Not Properly Multinational Study centers. A total of 1,586 patients from seven sites (five in the U.S., one in France, one in Norway) were enrolled from April 1999 to December 2000. To be eligible for the study, the patient had to have shortness of breath as their chief complaint. Patients under age 18 and whose dyspnea was clearly not secondary to CHF (e.g., trauma victims) were excluded. Patients with acute myocardial infarction, severe renal failure, and unstable angina were excluded.
Once a patient was identified as having dyspnea, written consent was obtained, and a blood sample was collected for purposes of measuring the BNP concentration. Other data were collected, including elements from the present and past history, the physical examination, and reports of other blood tests, interpretations of chest X-rays, or interpretations of other diagnostic tests. Echocardiograms were strongly encouraged, either in the emergency department (ED), as an outpatient, or in the hospital if the patient was admitted.
For each patient enrolled in the study, physicians assigned to the ED who were blinded to the results of BNP measurements made an assessment of the probability of the patient having CHF (0 to 100% clinical certainty) as the cause of his or her symptoms at the time of ED disposition.
Confirmation of the diagnosis
To determine the patient’s actual diagnosis, two cardiologists reviewed all medical records pertaining to the patient and made independent initial assessments of the final diagnosis: 1) CHF; 2) history of CHF but acute dyspnea due to non-cardiac cause; or 3) not CHF. The cardiologists were presented the components and summary of the Framingham (two major or one major and two minor criteria) CHF score and the National Health and Nutrition and Examination Survey (score ≥3) CHF score, calculated from the case report form. The cardiologists were blinded to the BNP results as well as the ED physicians’ diagnosis. They did have access to the ED data sheets and any additional information that became available after the ED visit. This included the following: official reading of the chest X-ray that was done in the ED by a radiologist; past medical history obtained from a medical chart that was not available at the time to the ED physicians; the results of subsequent tests such as echocardiography, radionuclide angiography, or left ventriculography done at the time of cardiac catheterization; and the hospital course for patients admitted to the hospital. For patients with a diagnosis other than CHF, confirmation was attempted using the following variables: normal chest X-ray (lack of heart enlargement and pulmonary venous hypertension); X-ray signs of chronic obstructive lung disease, pneumonia, or lung cancer; normal heart function by echocardiography, nuclear medicine ejection fractions (EF), or left ventriculography done at cardiac catheterization; abnormal pulmonary function tests or follow-up in pulmonary clinic; response to treatment in the ED or hospital with nebulizers, steroids, or antibiotics; and no CHF admissions over the next 30 days. In all cases of CHF, the two cardiologists were asked to agree on the degree of severity of CHF by ranking each patient as NYHA class I to IV.
Non-systolic versus systolic dysfunction
A subset of 452 patients with a final adjudicated diagnosis of CHF underwent echocardiography within 30 days of their visit to the ED. Patients with heart failure were defined as having systolic heart failure (S-CHF) if the ejection was 45% or less. Patients were defined as having non-systolic heart failure (NS-CHF) if EF was >45%.
Measurement of BNP plasma levels
During initial evaluations, a blood sample was collected into tubes containing potassium ethylenediaminetetraacetic acid. The BNP was measured in triplicate using the Triage B-Type Natriuretic Peptide test (Biosite Inc., San Diego, California). The Triage BNP test is a fluorescence immunoassay for the quantitative determination of BNP in whole blood and plasma specimens. The precision, analytical sensitivity, and stability characteristics of the system have been previously described (16,17). Triplicate BNP values were determined on site using the Triage BNP test with either whole blood or plasma samples.
For each of the different clinical and X-ray findings identified by ED physicians, the percentage of CHF cases with systolic dysfunction was computed. Group comparisons of BNP values were made using Mann-Whitney Utests. Other group comparisons were made using chi-squared tests and ttests for independent samples. Receiver operating characteristic (ROC) curves were used to evaluate the utility of BNP for various diagnostic comparisons. Sensitivity, specificity, and accuracy are reported for cut points of selected BNP concentrations. We also used a stepwise multivariate logistic model combining clinical findings and BNP values to differentiate between S-CHF and NS-CHF. The BNP values were log-transformed in this analysis to normalize the distribution.
The baseline characteristics for the overall study group of 1,586 patients are shown in Table 1. The mean age was 64 years. There were 883 (56%) males and 703 (44%) females. Ethnic makeup included 773 (49%) whites, 715 (45%) blacks, and 98 (6%) classified as other races. Baseline characteristics in patients with CHF who received echos were similar to those who did not receive echos except they were slightly younger, more often black, and had more fatigue.
On examination by ED physicians, 7% of patients had an S-3 gallop, 43% had rales in lower lung fields, 22% had jugular venous distention, and 42% had lower extremity edema. The final diagnosis (made retrospectively by two cardiologists) was CHF in 744 patients (47%), a history of CHF but dyspnea due to non-cardiac causes in 72 patients (4.5%), and no CHF in 770 patients (49%). In 97% of patients with CHF, the final diagnosis of CHF was confirmed by other tests (chest X-ray 79%, echocardiography 77%, radionuclide EF 15%, cardiac catheterization 19%, and response to treatment 86%).
Of the 452 patients with a final diagnosis of CHF who had echos within 30 days, 165 (36.5%) had preserved LV function on echocardiography (NS-CHF), whereas 287 had systolic dysfunction (S-CHF). The mean EF were 28 ± 0.6% and 59 ± 0.6% in systolic and non-systolic dysfunction, respectively (p < 0.001). Figure 1presents box plots showing median levels of BNP measured in the ED in patients with dyspnea not due to heart failure, and those with an adjudicated final diagnosis of heart failure, subdivided by those with S-CHF and those with NS-CHF. The median BNP level in the non-CHF groups was 34 pg/ml, significantly lower than either those with S-CHF or NS-CHF (p < 0.001 in both cases). Patients with NS-CHF had significantly lower BNP levels than those with S-CHF (413 pg/ml vs. 821 pg/ml, p < 0.001).
The relative proportions of systolic and non-systolic dysfunction differed significantly as a function of severity of CHF as agreed on by both cardiologists (p < 0.001). Figure 2shows a breakdown of each NYHA class in relation to S-CHF and NS-CHF. As the severity of heart failure worsened by NYHA classification, the proportion of S-CHF increased and the proportion of NS-CHF decreased.
Figure 3presents box plots showing median levels of BNP measured in men and women over 70 years of age with dyspnea not due to heart failure, and those with an adjudicated final diagnosis of heart failure, subdivided by those with S-CHF and those with NS-CHF. The difference between older men and woman with regard to BNP levels in the non-CHF group was not statistically significant. Among males, BNP was higher in both CHF groups than control (p < 0.001) but S-CHF and NS-CHF did not differ. Among females, BNP was higher than control in both types of CHF (p < 0.001) and women with systolic CHF had higher BNP levels than non-systolic failure patients (p = 0.03).
Table 2shows univariate differences in patients with S-CHF versus NS-CHF on presentation to the emergency room. Patients with NS-CHF less often had a history of heart failure (32% vs. 68%, p = 0.015) or myocardial infarction (29% vs. 71%, p = 0.002) than those with S-CHF. On physical examination, patients with NS-CHF had slower heart rates (88 vs. 93 beats/min, p = 0.028), higher systolic pressures (147 mm Hg vs. 137 mm Hg, p < 0.001), and lower mean oxygen saturations (91.8% vs. 94.2%, p = 0.002). A third heart sound was present less often in NS-CHF than S-CHF (26% vs. 74%, p = 0.045).
The ability of BNP to detect abnormal heart function was assessed with ROC curve analysis (Fig. 4). B-type natriuretic peptide was accurate in separating all CHF from non-CHF patients (area under the curve [AUC] = 0.90) with 90% sensitivity at the established cutoff of 100 pg/ml. The BNP levels were not very accurate in separating S-CHF from NS-CHF (AUC = 0.66, p < 0.001). Although a cut point of 100 pg/ml was 95% sensitive for detecting S-CHF, there was significant overlap between the two groups, with 86% of NS-CHF patients falling above this cutoff.
Logistic regression was used in a multivariate approach for differentiating systolic from non-systolic dysfunction in patients diagnosed with heart failure (Table 3). The BNP level, oxygen saturation, history of myocardial infarction, and heart rate were the variables most closely associated with separating the two entities.
As many as 40% to 55% of patients with the diagnosis of heart failure have preserved systolic function (1,2). The prevalence of CHF with preserved systolic function increases with age, with an approximate incidence of 15% to 25% in people less than 60 years old, 35% to 40% between 60 and 70 years, and 50% in people over 70 years of age (18,19).
Few data exist as to the proportion of patients who arrive at the ED with dyspnea as a result of heart failure with non-systolic LV dysfunction. The present study found that slightly over one-third of patients presenting to the ED had non-systolic dysfunction as determined by echocardiography done within 30 days of their index visit. Correctly diagnosing CHF in patients presenting with acute dyspnea will not only allow one to accrue the benefits of improved survival and increased well-being on medications such as angiotensin-converting enzyme inhibitors and beta-blockers (20), but also to avoid a misdiagnosis that could place the patient at risk for both morbidity and mortality (21).
Using a rapid assay for BNP has allowed the prompt differentiation of CHF from non-CHF causes of dyspnea, regardless of the type of heart failure (8,15,16). Additionally, BNP levels are highly prognostic in this setting. Harrison et al. (22)followed 325 patients for six months after an index visit to the ED for dyspnea and found that the relative risk of six-month CHF death in patients with BNP levels more than 230 pg/ml was 24.
It has been previously demonstrated that in the non-emergency setting BNP is elevated in patients with NS-CHF (13,14,20,23–25). Although BNP levels could not by themselves differentiate between S-CHF and NS-CHF, a low BNP level in the setting of normal systolic function by echocardiography was able to rule out clinically significant diastolic abnormalities seen on echo. On the other hand, elevated BNP levels in patients with normal systolic function, especially in older patients with a history of CHF, correlated to ventricular filling abnormalities on Doppler studies (20,24).
In the present study, BNP was elevated in NS-CHF to a median value of 413 pg/ml, higher than that seen in most ambulatory patients with NS-CHF (13,14). Interestingly, Lubien et al. (25)found BNP levels of 408 pg/ml in patients with the restrictive filling pattern, the mitral Doppler pattern most often associated with elevated LV end-diastolic pressures. In patients with normal EF, a high BNP level usually meant diastolic dysfunction. In the ED setting this may be true as well. In the setting of dyspnea in which LV function is normal, a BNP level <100 pg/ml gave a negative predictive value of 96%.
With few exceptions, NS-CHF has not been shown to be distinguishable from S-CHF solely on the basis of history, physical examination, chest X-ray, and electrocardiogram (1–3,26). Recently, Thomas et al. collected data on 225 patients hospitalized with CHF and found that differences in clinical parameters could not predict systolic function in these patients (26); the investigators suggested that specialized tests of ventricular function were needed. The present study supports these data in that of all historical and physical examination variables, only an absent history of myocardial infarction or heart failure, an absent third heart sound, and a high systolic blood pressure and lower heart rate were more predictive of NS-CHF than S-CHF. B-type natriuretic peptide levels are higher in patients with S-CHF than in those with NS-CHF, possibly reflecting an association with greater pathology in patients presenting with S-CHF, as reflected by their higher NYHA classifications. As the BNP level rises, the positive predictive value for S-CHF also increases, so that a BNP level >400 pg/ml offers a 72% likelihood that a patient will have S-CHF. However, the marked overlap in BNP values clearly limits its usefulness in separating the two groups in the clinical setting.
The importance of differentiating patients with NS-CHF from those with S-CHF lies in the underlying etiology and the subsequent treatment of the individual patient. Abnormal NS-CHF can be precipitated by ischemia, abnormally high blood pressure, or atrial fibrillation, all of which would receive different workups and treatment than used for patients with S-CHF, who are more likely to be admitted to the hospital and receive parenteral vasodilator and/or inotropic therapy. The BNP levels clearly cannot substitute for measurements of LV function and should not be considered a surrogate for echocardiography.
In the future, BNP levels may provide a surrogate end point for the evaluation of various treatments of heart failure. Falling BNP levels with treatment is associated with falling wedge pressures, a lower readmission rate to the hospital, and a better prognosis (21,27). Thus, monitoring BNP levels in future treatment protocols for NS-CHF may provide valuable information regarding drug efficacy and patient outcomes.
Echocardiographic recordings form the basis for differentiating between S-CHF and NS-CHF in the current study. Only 710 of the 1,586 patients (45%) enrolled in the Breathing Not Properly multinational study received echocardiograms within 30 days of their ED visit. Numerous previous reports have validated the ability of cardiac ultrasonography to detect abnormalities of contractile function and to quantitate LV volumes and EF (28,29). All patients in this study so designated had clear-cut evidence of LV systolic dysfunction. Although diastolic dysfunction implies an abnormal relationship between LV volume and pressure, echocardiography is capable of assessing only parameters related to volume. As Doppler parameters such as transmitral and pulmonary venous flow velocities were not always recorded and provide only indirect measurements of diastolic performance, they were not evaluated in the context of the present study. We also used a cutoff of 30 days for echocardiography, which meant that the study might not have reflected the actual status in the ED (30).
We conclude that BNP is a useful test to distinguish patients with heart failure from those without heart failure. This utility extends to the group with NS-CHF, in whom the diagnosis is more difficult and often incorrectly excluded when the presence of normal LV systolic function is known. However, BNP cannot reliably distinguish S-CHF from NS-CHF. Thus, a measurement of LV function is required to make this distinction and guide therapy accordingly.
The authors are indebted to the efforts of the ED staff at the following BNP Multinational Study Centers: San Diego Veterans Affairs Medical Center, San Diego, CA; Henry Ford Hospital, Detroit, MI; Hospital of the University of Pennsylvania, Philadelphia, PA; Hospital Bichat, Paris, France; Ullevål University Hospital, Oslo, Norway; University of Cincinnati Medical Center, Cincinnati, OH; and Hartford Hospital, Hartford, CT.
Breathing Not Properly Multinational Study Investigators:University of California, San Diego, Veterans Affairs Medical Center, San Diego, CA: Study Principal Investigator: Alan S. Maisel, MD; Site Principal Investigators: Radmila Kazanegra, MD, Padma Krishnaswamy, MD, Patricia Hlavin, MD, Leslie A. Lenert, MD, Padma Krishnaswamy, MD; Biostatistician: Paul Clopton, MS; Henry Ford Hospital, Detroit, MI: Site Principal Investigators: Richard M. Nowak, MD, MBA, James McCord, MD; Study Coordinators: Michele Whitican, RN, James Babiarz, RN; University of Pennsylvania, Philadelphia, PA: Site Principal Investigators: Judd E. Hollander, MD, Howard C. Herrmann, MD, Evan Loh, MD; Study Coordinator: Frank D. Sites, RN, BSN; Hopital Bichat, Paris, France: Site Principal Investigators: Philippe Duc, MD, Gabriel Steg, MD; Co-Investigators: Marie Claude Aumont, MD, Valerie Beaumesnil, MD, Lamia Hafi, MD, Armelle Desplanques, MD, Joelle Benessiano, MD; Ullevål University Hospital, Oslo, Norway; Site Principal Investigators: Arne Westheim, MD, PhD, Torbjørn Omland, MD, PhD, MPH, Cathrine Wold Knudsen, MD; Co-Investigators: Alexandra Finsen, MD, Jon Sigurd Riis, MD, Tor Ole Klemsdal MD, PhD; University of Cincinnati College of Medicine, Cincinnati, OH: Site Principal Investigators: Alan B. Storrow, MD; University of Kentucky College of Medicine, Lexington, KY: Co-Investigators: Sumant Lamba, MD, William T. Abraham, MD; Hartford Hospital, Hartford, CT: Site Principal Investigators: Alan H. B. Wu, PhD, Alberto Perez, MD; University of Missouri-Kansas City School of Medicine, Truman Medical Center, Kansas City, MO: Study Co-Principal Investigator: Peter A. McCullough, MD, MPH.
Funding/Support:Triage BNP devices and meters and some financial support were provided by Biosite, Incorporated, San Diego, CA.
The following authors have financial disclosures relating to Biosite, Inc.: Alan Maisel, MD—consultant and has received research support; Peter McCullough, MD—consultant and has received research support; Alan Wu, PhD— consultant and has received research support; Richard Nowack, MD—consultant and has received research support; James McCord, MD—consultant and has received research support.
☆ Triage B-Type Natriuretic Peptide devices and meters and some financial support were provided by Biosite, Inc., San Diego, California. Lynne Warner-Stevenson, MD, was the Guest Editor for this paper
Presented in part at the 51st Scientific Sessions of the American College of Cardiology, Atlanta, Georgia, March 17, 2002.
- area under the curve
- B-type natriuretic peptide
- congestive heart failure
- emergency department
- ejection fraction
- left ventricular
- non-systolic congestive heart failure
- New York Heart Association
- receiver operating characteristic
- systolic congestive heart failure
- Received June 1, 2002.
- Revision received October 21, 2002.
- Accepted November 19, 2002.
- American College of Cardiology Foundation
- Khandelwal A.,
- McKinnon J.E.,
- Shenkman H.J.,
- et al.
- Clerico A.,
- Iervasi G.,
- Chicca M.,
- et al.
- Dao Q.,
- Krishnaswamy P.,
- Kazanegra R.,
- et al.
- Richards AM, Nicholls MG, Yandle TG, et al. Plasma N-terminal pro-brain natriuretic peptide and adrenomedullin. Circulation 1998;97:1921–9
- Tsutamoto T.,
- Wada A.,
- Maeda K.,
- et al.
- Cowie M.R.,
- Struthers A.D.,
- Wood D.A.,
- et al.
- Yu C.M.,
- Sanderson J.E.,
- Shum I.O.L.,
- et al.
- ↵Maisel AM, Krishnaswamy P, Nowak R, et al. Bedside B-type natriuretic peptide in the emergency diagnosis of heart failure: primary results from the Breathing Not Properly (BNP) Multinational study. Presented at ACC meeting, Atlanta, March 17–20, 2002
- Morrison K.L.,
- Harrison A.,
- Krishnaswamy P.,
- Kazanegra R.,
- Clopton P.,
- Maisel A.S.
- Cheng V.L.,
- Krishnaswamy P.,
- Kazanegra R.,
- Garcia A.,
- Gardetto N.,
- Maisel A.S.
- Lubien E.,
- DeMaria A.,
- Krishnaswamy P.,
- et al.
- Sodian R.,
- Loebe M.,
- Schmitt C.,
- et al.
- ↵Remme WJ, Swedberg K. Guidelines for the diagnosis of treatment of chronic heart failure. Eur Heart J 2001;22:1527–60
- Schiller N.B.,
- Acquatella H.,
- Ports T.A.,
- et al.