1muel.oa

Use of B-Type Natriuretic Peptide in the Evaluation Christian Mueller, M.D., André Scholer, Ph.D., Kirsten Laule-Kilian, B.Sc., Benedict Martina, M.D., Christian Schindler, Ph.D., Peter Buser, M.D., Matthias Pfisterer, M.D., b a c k g r o u n d
B-type natriuretic peptide levels are higher in patients with congestive heart failure From the Department of Internal Medicine, Medical Division A (C.M., K.L.-K., A.P.P.), than in patients with dyspnea from other causes.
the Department of Laboratory Medicine(A.S.), the Emergency Department (B.M.), the Institute for Social and PreventiveMedicine (C.S.), and the Division of Cardi- We conducted a prospective, randomized, controlled study of 452 patients who pre- ology (P.B., M.P.), University of Basel, sented to the emergency department with acute dyspnea: 225 patients were randomly University Hospital, Basel, Switzerland.
assigned to a diagnostic strategy involving the measurement of B-type natriuretic pep- Address reprint requests to Dr. Mueller at Medizinische Klinik A, Universitätsklinik, tide levels with the use of a rapid bedside assay, and 227 were assessed in a standard Petersgraben 4, CH-4031 Basel, Switzer- manner. The time to discharge and the total cost of treatment were the primary end land, or at chmueller@uhbs.ch.
points.
N Engl J Med 2004;350:647-54.
Copyright 2004 Massachusetts Medical Society. Base-line demographic and clinical characteristics were well matched between the twogroups. The use of B-type natriuretic peptide levels reduced the need for hospitaliza-tion and intensive care; 75 percent of patients in the B-type natriuretic peptide groupwere hospitalized, as compared with 85 percent of patients in the control group(P=0.008), and 15 percent of those in the B-type natriuretic peptide group required in-tensive care, as compared with 24 percent of those in the control group (P=0.01). Themedian time to discharge was 8.0 days in the B-type natriuretic peptide group and 11.0days in the control group (P=0.001). The mean total cost of treatment was $5,410 (95percent confidence interval, $4,516 to $6,304) in the B-type natriuretic peptide group,as compared with $7,264 (95 percent confidence interval, $6,301 to $8,227) in the con-trol group (P=0.006). The respective 30-day mortality rates were 10 percent and 12percent (P=0.45).
c o n c l u s i o n s
Used in conjunction with other clinical information, rapid measurement of B-type natri-uretic peptide in the emergency department improved the evaluation and treatment ofpatients with acute dyspnea and thereby reduced the time to discharge and the totalcost of treatment.
Downloaded from www.nejm.org at UNIVERSITY OF BASEL on November 10, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine hmillion patients have heart failure in setting and study population
North America and Europe, with nearly 1.5 million The B-Type Natriuretic Peptide for Acute Shortnessnew cases every year.1-5 Heart failure is the most of Breath Evaluation (BASEL) Study was a prospec-frequent cause of hospitalization among people tive, randomized, controlled, single-blind study con-older than 65 years of age, and these hospitaliza- ducted in the emergency department of the Univer-tions are an important part of the enormous cost of sity Hospital in Basel, Switzerland. Patients werethe disease. Over the past decade, the rate of hos- evaluated in the emergency department by at leastpitalization for heart failure has increased by 159 two physicians: a resident in internal medicine andpercent.3 It is estimated that in the United States in an internal-medicine specialist. The study investi-2001, the direct cost of the care of patients with gators were not directly involved in patient care inheart failure exceeded $24 billion.3 Therefore, cost- the emergency department, nor did they have any in-effective management is of paramount importance. fluence on the decision to discharge patients fromHowever, the rapid and accurate differentiation of the ward. The study was carried out according toheart failure from other causes of dyspnea remains the principles of the Declaration of Helsinki anda clinical challenge, especially in the emergency de- approved by the local ethics committee. Written in-partment.4-10 After evaluating a patient’s symptoms, formed consent was obtained from all participat-conducting a physical examination, and perform- ing patients.
ing electrocardiography and chest radiography, the We screened 665 consecutive adults who pre- clinician is often left with considerable diagnostic sented to the emergency department between Mayuncertainty, which results in misdiagnosis and de- 2001 and April 2002. Eligible patients were thoselays the initiation of appropriate therapy.7,10 In addi- who had acute dyspnea as the primary symptom,tion, the misdiagnosis of heart failure causes mor- with no obvious traumatic cause of dyspnea. Pa-bidity and increases the time to discharge and the tients with severe renal disease (defined by a serumcost of treatment, because the use of a treatment creatinine level of more than 250 µmol per literstrategy for other conditions, such as chronic ob- [2.8 mg per deciliter]), patients with cardiogenicstructive pulmonary disease, may be hazardous to shock, and patients who requested an early trans-patients with heart failure, and vice versa.6,8,9 fer to another hospital were excluded. There were Observational studies have suggested that, when no limitations to entry according to the time of day used in conjunction with other clinical information, at which patients arrived in the emergency depart-B-type natriuretic peptide levels may be useful in ment or the availability of research staff.
establishing or ruling out the diagnosis of heart fail- A total of 452 patients were enrolled in the trial, ure in patients with acute dyspnea.6,7,10-14 B-type and group assignment was accomplished with the
natriuretic peptide is a 32-amino-acid polypep- use of a computer-generated randomization scheme
tide secreted by the cardiac ventricles in response in a 1:1 ratio without stratification. A total of 225
to ventricular volume expansion and pressure over- patients were randomly assigned to be evaluated
load.15-17 The levels of B-type natriuretic peptide with the use of a diagnostic strategy that included
are elevated in patients with left ventricular dysfunc- the rapid bedside measurement of B-type natriuret-
tion, and the levels correlate with both the severity ic peptide levels, and 227 were assigned to be eval-
of symptoms and the prognosis.3,6,7,9-19 However, uated with the use of the conventional diagnostic
the clinical effect of this diagnostic test on the eval- strategy. The B-type natriuretic peptide was not mea-
uation and treatment, outcome, and cost of treat- sured for clinical purposes by clinicians who treat-
ment of patients with dyspnea is unknown. There- ed patients in the control group, nor was it mea-
fore, we performed a randomized, controlled trial to sured serially in either of the groups.
test the hypothesis that a diagnostic strategy guid-
ed by the rapid measurement of B-type natriuretic routine clinical assessment
peptide levels would improve the evaluation and All patients underwent an initial clinical assess-
care of patients with acute dyspnea who present to ment that, in general, included a clinical history tak-
the emergency department and would thereby re- ing, a physical examination, electrocardiography,
duce the time to discharge and the total cost of pulse oximetry, blood tests, and chest radiography.
treatment.
Echocardiography and pulmonary-function tests Downloaded from www.nejm.org at UNIVERSITY OF BASEL on November 10, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. b - t y p e n a t r i u r e t i c p e p t i d e i n a c u t e d y s p n e a were strongly recommended on an outpatient ba- in the control group were evaluated and treated ac-sis for patients who were released from the emer- cording to the most recent clinical guidelines.4,5gency department, as well as for the patients whowere admitted.
e n d p o i n t s
The time to discharge and the cost of treatment m e a s u r e m e n t a n d i n t e r p r e t a t i o n
were the primary end points of the study. Second- o f b - t y p e n a t r i u r e t i c p e p t i d e l e v e l s
ary end points included in-hospital and 30-day During the initial evaluation, at the time of veni- mortality. The time to discharge was defined as thepuncture for routine blood tests, a 5-ml specimen interval from presentation at the emergency depart-of venous blood was collected in tubes containing ment to discharge. Patients who died in the hospi-potassium EDTA. During a 15-minute period, B-type tal were excluded from the calculation of this endnatriuretic peptide was measured with the use of point. Since ratios of costs to charges have not beena rapid fluorescence immunoassay (Biosite Diag- defined for the majority of services and departmentsnostics). The precision, analytic sensitivity, and sta- at our institution, hospital charges were used asbility of the system have been described previous- the most appropriate estimate of the true costs.21,22ly.14,18 In brief, the coefficient of variation within a To avoid an imbalance owing to differences in re-given assay has been reported to be 9.5 percent, imbursement or charges associated with different12.0 percent, and 13.9 percent for levels of 28.8, types or classes of insurance, charges were stan-584.0, and 1180.0 pg per milliliter, respectively, and dardized according to the actual rates for patientsthe coefficient of variation among assays is known with general insurance who were living in Basel. Theto be 10.0 percent, 12.4 percent, and 14.8 percent, current reimbursement for the measurement ofrespectively.14,18 The limit of analytic sensitivity was B-type natriuretic peptide in Switzerland ($47) wasless than 5.0 pg per milliliter, with a measurable used. The time to treatment was defined as the in-range of 0 to 1300 pg per milliliter.
terval from presentation to the initiation of the ap- In the group in which B-type natriuretic peptide propriate therapy — other than bed rest and supple- levels were measured, diagnostic and therapeutic mental oxygen — according to the final discharge
decisions were not based on the B-type natriuretic diagnosis. This therapy included diuretics or vaso-
peptide levels alone; instead, this information was dilators in patients with heart failure, anticoagulants
considered in the context of the other clinical infor- in patients with pulmonary embolism, and inhaled
mation obtained and the physicians’ clinical im- bronchodilators or systemic corticosteroids in pa-
pressions, as previously described.20 In brief, we tients with an exacerbation of obstructive pulmo-
used a B-type natriuretic peptide level of 100 pg per nary disease. All end points were assessed in a blind-
milliliter to separate dyspnea caused by heart fail- ed fashion by physicians who were not involved in
ure from other causes of dyspnea.3,6,7,9-13,20 In pa- patient care, with the use of all medical records per-
tients with a B-type natriuretic peptide level below taining to each patient.
100 pg per milliliter, the diagnosis of heart failure
was considered unlikely, and alternative causes of statistical analysis
dyspnea had to be investigated. In patients with a The statistical analyses were performed with the
B-type natriuretic peptide level of more than 500 pg use of the SPSS/PC software package (version 11.0,
per milliliter, heart failure was considered the most SPSS). A P value of less than 0.05 was considered to
likely diagnosis, and rapid therapy with diuretics, indicate statistical significance. All data were ana-
nitroglycerin, angiotensin-converting–enzyme in- lyzed according to the intention-to-treat principle.
hibitors, and morphine was recommended. For pa- Comparisons were made with the use of the t-test,
tients with B-type natriuretic peptide levels between the Mann–Whitney U test, Fisher’s exact test, or the
100 and 500 pg per milliliter, the protocol recom- chi-square test, as appropriate. All hypothesis test-
mended the use of clinical judgment and possible ing was two-tailed. The trial was designed to enroll
further diagnostic testing to rule out stable base- 222 patients in each group. This number provided
line left ventricular dysfunction and other condi- the study with a power of 80 percent to detect a re-
tions as the real cause of acute dyspnea.3,6,7,9-14,20 duction in the time to discharge from 10.0 to 8.0
No formal adjustment was recommended regard- days (20 percent) with the use of the diagnostic
ing the B-type natriuretic peptide cutoff values in strategy guided by measurement of the B-type na-
patients with mild chronic kidney disease. Patients triuretic peptide level. Assumptions included the
Downloaded from www.nejm.org at UNIVERSITY OF BASEL on November 10, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine Table 1. Base-Line Characteristics of the Patients.
B-Type Natriuretic Peptide Group
Control Group
Characteristic
Downloaded from www.nejm.org at UNIVERSITY OF BASEL on November 10, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. b - t y p e n a t r i u r e t i c p e p t i d e i n a c u t e d y s p n e a Table 1. (Continued.)
B-Type Natriuretic Peptide Group
Control Group
Characteristic
* Four patients in the B-type natriuretic peptide group and two patients in the control group had shortness of breath only † To convert values for creatinine to milligrams per deciliter, divide by 88.4.
‡ ACE denotes angiotensin-converting–enzyme inhibitor or angiotensin-receptor blocker.
use of a two-tailed test, a 5 percent level of signifi- patients were women. The medical history includedcance, and a standard deviation of 7.5 days in both coronary artery disease in 50 percent of patients, hy-groups.
pertension in 52 percent, chronic obstructive pul-monary disease in 31 percent, any pulmonary dis-ease in 50 percent, and diabetes in 23 percent.
The median time from presentation at the emer- A total of 452 patients were enrolled. The base-line gency department to the initiation of the appropri-characteristics were well matched between the ate therapy according to the final discharge diag-study groups (Table 1). The mean age was 71 years. nosis was 90 minutes in the control group and 63In both groups, slightly more than 40 percent of the minutes in the B-type natriuretic peptide group Downloaded from www.nejm.org at UNIVERSITY OF BASEL on November 10, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine (P=0.03) (Table 2). The use of B-type natriuretic Heart failure was the final discharge diagnosis peptide levels significantly reduced the need for in 45 percent of patients in the B-type natriuretichospitalization and intensive care: 75 percent of pa- peptide group and 51 percent of patients in the con-tients in the B-type natriuretic peptide group were trol group (P=0.2). Exacerbation of obstructive pul-hospitalized, as compared with 85 percent of those monary disease was more commonly the cause ofin the control group (P=0.008), and 15 percent of acute dyspnea in the B-type natriuretic peptidepatients in the B-type natriuretic peptide group re- group than in the control group (23 percent vs. 11quired intensive care, as compared with 24 percent percent, P=0.001). The treating physician indicat-of those in the control group (P=0.01).
ed that two causes contributed to the acute dyspnea There was a considerable range in the time to in 11 patients (5 percent) in the B-type natriuretic discharge, reflecting the variety of diseases respon- peptide group and in 10 patients (4 percent) in thesible for acute dyspnea. As shown in Figure 1, the control group (P=0.81).
time to discharge was significantly shorter in the Clinical 30-day follow-up data were available for B-type natriuretic peptide group (median, 8.0 days) all patients. The rates of readmission and mortalitythan in the control group (median, 11.0 days; P= within 30 days after discharge were similarly low in 0.001). This difference translated into a significant the two groups (Table 2). The 30-day mortality rate difference in the mean total cost of treatment: was 10 percent in the B-type natriuretic peptide$7,264 in the control group, as compared with group and 12 percent in the control group. Among$5,410 in the B-type natriuretic peptide group (P= patients who were not initially admitted, rates of sec-0.006). Twenty-one patients (9 percent) in the con- ondary admission were 5 percent in the B-type natri-trol group died in the hospital, as compared with uretic peptide group (3 of 56 patients) and 9 per-13 patients in the B-type natriuretic peptide group cent in the control group (3 of 34 patients, P=0.67),(6 percent, P=0.19).
and the respective 30-day mortality rates were 4 per-cent (2 deaths) and 3 percent (1 death) (P=1.00).
Table 2. End Points.*
B-Type Natriuretic
This randomized, controlled trial examined the ef- Peptide Group
Control Group
End Point
fect of the measurement of B-type natriuretic pep-tide levels in the emergency diagnosis of patients with acute dyspnea. The use of B-type natriuretic peptide levels in conjunction with other clinical in- formation reduced the time to the initiation of the Cumulative Percent
Time to Discharge (days)
* The time to treatment was defined as the interval from presentation at the Figure 1. Cumulative Frequency Distribution Curve
emergency department to the initiation of the appropriate therapy according for the Time to Discharge of Patients in the B-Type
Natriuretic Peptide Group as Compared with Those
† The Mann–Whitney U test was used.
in the Control Group.
Downloaded from www.nejm.org at UNIVERSITY OF BASEL on November 10, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. b - t y p e n a t r i u r e t i c p e p t i d e i n a c u t e d y s p n e a most appropriate therapy, the need for hospitaliza- considerable number of patients in the controltion and intensive care, the time to discharge, and group. Our findings, along with those of other in-the total cost of treatment. Given the morbidity vestigators, provide support for the inclusion ofassociated with acute dyspnea and the cost associ- the measurement of natriuretic peptides in the re-ated with heart failure, chronic obstructive pulmo- cent European guidelines for the diagnosis of heartnary disease, and other disorders that cause dys- failure.5pnea,1-5,8,9,14 B-type natriuretic peptide testing is A particular strength of our study is that the likely to be of value in the treatment of patients study population was highly representative of thewith acute dyspnea. The mean total cost of treat- elderly population of patients with heart failure inment in this study was similar to the expenditures clinical practice.1-3 The mean age was 71 years,in the United States. In 1997, an estimated $5,501 nearly half the patients were women, and coexist-was spent for every hospital-discharge diagnosis of ing conditions were common. The rapid and accu-heart failure.3 rate differentiation of heart failure from other caus- B-type natriuretic peptide testing reduced the es of acute dyspnea in such patients is often difficult, total cost of treatment by 26 percent. This finding although essential for cost-effective management.
is supported by a retrospective analysis of the cost The symptoms and signs of heart failure are nei-effectiveness of the use of B-type natriuretic pep- ther sensitive nor specific and considerably overlaptide levels in screening for left ventricular systolic those of pulmonary disease.4-7,13 The approach todysfunction in the general population, which also the emergency diagnosis of acute dyspnea has beenshowed a 26 percent reduction in cost.23 fundamentally unchanged for decades and has been Our findings extend the conclusions of observa- complemented by electrocardiography, chest radi- tional studies in which the use of the measurement ography, and echocardiography for the assessmentof B-type natriuretic peptide levels was validated by of left ventricular function. Unfortunately, thesecomparison with a retrospectively adjudicated di- methods have important limitations.25-27agnosis of heart failure by independent cardiol- The clinical experience with B-type natriuretic ogists.3,6,7,9-14 In the largest of these studies — peptide testing is limited. Our interpretation of thethe Breathing Not Properly Multinational Study test results was based on the data available when— B-type natriuretic peptide levels by themselves the study protocol was devised. Further studieswere more accurate than any historical or physical should help to optimize the use of B-type natriureticfinding or laboratory value in identifying heart fail- peptide measurements in clinical practice. The useure as the cause of dyspnea. The diagnostic accura- of normal values corrected for age and sex may rep-cy of B-type natriuretic peptide at a cutoff value of resent a clinically significant advance, since the lev-100 pg per milliliter was 83 percent, with a sensitiv- els of B-type natriuretic peptide increase with ageity of 90 percent and a specificity of 76 percent.6,7 and are higher in women than in men.28 Moreover, In our study, exacerbation of obstructive pulmo- a heart-failure diagnosis nomogram has been devel- nary disease was more often the cause of acute dys- oped.8 In patients with severe renal disease, B-typepnea in the B-type natriuretic peptide group than in natriuretic peptide levels are increased. The meanthe control group. This finding corresponds well B-type natriuretic peptide level in patients with awith a recent observation that exacerbation of chron- noncardiac cause of dyspnea and an estimated glo-ic obstructive pulmonary disease frequently escapes merular filtration rate of less than 60 ml per minuterecognition in the emergency department24 and is per 1.73 m2 of body-surface area was nearly 300 pgalso in agreement with the high negative predic- per milliliter in the Breathing Not Properly Multi-tive value of the B-type natriuretic peptide level for national Study.29 Therefore, higher cutoff valuesthe diagnosis of heart failure. B-type natriuretic need to be identified for this important patientpeptide levels below 100 pg per milliliter in a pa- population.
tient with acute dyspnea make the diagnosis of In conclusion, we found that when used in con- heart failure very unlikely and apparently help cli- junction with other clinical information, rapid mea-nicians focus on the most common alternative di- surement of B-type natriuretic peptide levels in theagnosis. Obstructive pulmonary disease was present emergency department improves the care of pa-in one third of our patients and may well have gone tients with acute dyspnea and thereby reduces theunrecognized as the cause of acute dyspnea in a time to discharge and the total cost of treatment.
Downloaded from www.nejm.org at UNIVERSITY OF BASEL on November 10, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. b - t y p e n a t r i u r e t i c p e p t i d e i n a c u t e d y s p n e a Supported by research grants from the Swiss National Science We are indebted to the emergency department staff at University Foundation, the Swiss Heart Foundation, the Novartis Foundation, Hospital Basel for their valuable efforts, to all participating patients, the Krokus Foundation, and the University of Basel (to Dr. Mueller).
and to Drs. Barbara Frana, Daniel Rodriguez, and Bruno Schurter Diagnostic devices and reagents (Triage) were provided by Biosite, for their help with data management.
r e f e r e n c e s
in differentiating congestive heart failure ment of patients with congestive heart fail- et al. Confirmation of a heart failure epi- ure? Swiss Med Wkly 2002;132:618-22.
21. Weinstein MC, Siegel JE, Gold MR,
(REACH) study. J Am Coll Cardiol 2002;39: 11. Davis M, Espiner E, Richards G, et al.
of the Panel on Cost-effectiveness in Health Plasma brain natriuretic peptide in assess- 22. Siegel JE, Weinstein MC, Russell LB,
12. McDonagh TA, Robb SD, Murdoch DR,
cost-effectiveness analysis. JAMA 1996;276: et al. Biochemical detection of left-ventricu- 2003 update. Dallas: American Heart Asso- lar systolic dysfunction. Lancet 1998;351:9- 23. Nielsen OW, McDonagh TA, Robb SD,
13. Dao Q, Krishnaswamy P, Kazanegra R,
cost-effectiveness of using plasma natri- ACC/AHA guidelines for the evaluation and et al. Utility of B-type natriuretic peptide in uretic peptide in screening for left ventricu- management of chronic heart failure in the the diagnosis of congestive heart failure in lar systolic dysfunction in the general popu- adult: executive summary: a report of the an urgent-care setting. J Am Coll Cardiol lation. J Am Coll Cardiol 2003;41:113-20.
24. Incalzi RA, Fuso L, Serra M, et al. Exac-
14. Cabanes L, Richaud-Thiriez B, Fulla Y,
erbated chronic obstructive pulmonary dis- et al. Brain natriuretic peptide blood levels ease: a frequently unrecognized condition.
Guidelines for the Evaluation and Manage- in the differential diagnosis of dyspnea.
ment of Heart Failure). Circulation 2001; 25. Rihal CS, Davis KB, Kennedy JW, Gersh
15. Yasue H, Yoshimura M, Sumida H, et al.
BJ. The utility of clinical, electrocardio- Localization and mechanism of secretion of graphic, and roentgenographic variables in ment of Chronic Heart Failure, European So- the prediction of left ventricular function.
ciety of Cardiology. Guidelines for the diag- with those of A-type natriuretic peptide in nosis and treatment of chronic heart failure.
normal subjects and patients with heart fail- 26. Gillespie ND, McNeill G, Pringle T,
Eur Heart J 2001;22:1527-60. [Erratum, Eur Ogston S, Struthers AD, Pringle SD. Cross 16. Hama N, Itoh H, Shirakami G, et al.
sectional study of contribution of clinical as- Rapid ventricular induction of brain natri- sessment and simple cardiac investigations RM, et al. Rapid measurement of B-type na- uretic peptide gene expression in experi- to diagnosis of left ventricular systolic dys- triuretic peptide in the emergency diagnosis mental acute myocardial infarction. Circula- of heart failure. N Engl J Med 2002;347:161- 17. Wiese S, Breyer T, Dragu A, et al. Gene
27. Zile MR, Gaasch WH, Carroll JD, et al.
expression of brain natriuretic peptide in Heart failure with a normal ejection frac- et al. B-type natriuretic peptide and clinical isolated atrial and ventricular human myo- tion: is measurement of diastolic function cardium: influence of angiotensin II and di- necessary to make the diagnosis of diastolic failure: analysis from Breathing Not Proper- astolic fiber length. Circulation 2000;102: heart failure? Circulation 2001;104:779-82.
ly (BNP) Multinational Study. Circulation 28. Redfield MM, Rodeheffer RJ, Jacobsen
18. Cheng V, Kazanegra R, Garcia A, et al.
SJ, Mahoney DW, Bailey KR, Burnett JC Jr.
A rapid bedside test for B-type peptide pre- Plasma brain natriuretic peptide concentra- gestive heart failure: prompt diagnosis may dicts treatment outcomes in patients admit- tion: impact of age and gender. J Am Coll ted for decompensated heart failure: a pilot study. J Am Coll Cardiol 2001;37:386-91.
29. McCullough PA, Duc P, Omland T, et al.
19. Koglin J, Pehlivanli S, Schwaiblmair M,
B-type natriuretic peptide and renal func- Vogeser M, Cremer P, von Scheidt W. Role of tion in the diagnosis of heart failure: an analy- length of stay in congestive heart failure.
brain natriuretic peptide in risk stratifica- sis from the Breathing Not Properly Multi- tion of patients with congestive heart fail- 10. Morrison LK, Harrison A, Krishna-
ure. J Am Coll Cardiol 2001;38:1934-41.
swamy P, Kazanegra R, Clopton P, Maisel A.
20. Mueller C, Buser P. B-type natriuretic
Copyright 2004 Massachusetts Medical Society. Utility of a rapid B-natriuretic peptide assay peptide (BNP): can it improve our manage- Downloaded from www.nejm.org at UNIVERSITY OF BASEL on November 10, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved.

Source: http://www.dyspnea.ch/Publikationen/PDF/032.pdf