- Email: [email protected]
Aminoterminal pro type B natriuretic peptide as a predictive and prognostic marker in patients with chronic heart failure
FAILING HEART—MEDICAL ASPECTS
Aminoterminal pro Type B Natriuretic Peptide as a Predictive and Prognostic Marker in Patients With Chronic Heart Failure Markus Rothenburger, MD,a Thomas Wichter, MD,b Christof Schmid, MD,a Jörg Stypmann, MD,b Tonny D. T. Tjan, MD,a Elmar Berendes, MD,c Christian Etz, MD,a Aurélien Pioux, MD,d Andreas Löher, MD,a Frauke Wenzelburger, MD,a Gabriele Drees, MD,a Andreas Hoffmeier, MD,a Günter Breithardt, MD,b and Hans Heinrich Scheld, MDa Background: B-type natriuretic peptide (BNP) is released from the cardiac ventricles in response to increased wall tension. We studied the relation of NT-proBNP to Heart Failure Survival Score (HFSS) and New York Heart Association (NYHA) class in patients with chronic heart failure (CHF). We also studied the impact for recipient selection for cardiac transplant and assessed it as a predictive and prognostic marker of CHF. Methods: A total of 550 patients with dilative cardiomyopathy (n ⫽ 323), and coronary artery disease (n ⫽ 227) were prospectively examined. All patients underwent spiroergometry, echocardiography, right heart catheterization, and electrocardiogram. Routine blood levels and NT-proBNP were measured. The clinical selection for cardiac transplant candidates was adjudicated by 2 independent cardiologists who were blinded to the results of NT-proBNP assays. Clinical outcome and predictive power of NT-proBNP were analyzed. Results: NT-proBNP levels in patients clinically considered for cardiac transplantation were significantly higher (2293 ng/ml vs 493 ng/ml; p ⬍ 0.001). The receiver operating characteristic (ROC) analysis regarding transplant candidacy showed an area under the ROC curve (AUC) of 0.84 ⫾ 0.01 for HFSS, 0.86 ⫾ 0.001 for NYHA, and 0.96 ⫾ 0.01 for NT-proBNP. Patients with increasing NT-proBNP levels or remaining elevated levels despite adequate heart insufficiency treatment were maintained with left ventricular assist device implantation (n ⫽ 10) or urgent heart transplantation (n ⫽ 2). Patients with NT-proBNP levels above 5000 pg/ml had a mortality rate of 28.4% per year. Twenty-eight patients died during the observation period; all these patients were within NYHA Classes 3 and 4 (NT-proBNP 5423 ⫾ 423 ng/ml). Conclusions: NT-proBNP discriminates patients at high likelihood of being a candidate for transplantation and provides prognostic informations in patients with CHF. NT-proBNP levels above 5000 pg/ml at admission were associated with death, and these levels markedly discriminated candidates for left ventricular assist devices or urgent transplantation. J Heart Lung Transplant.
An increasing number of patients with chronic heart failure (CHF) may have significant morbidity and mortality, with 65% of patients dying within 5 years.1 The health care expenditure for this disease is substantial, accounting for 1% to 2% of total health care costs, most of which is the result of repeated hospitalizations.2
From the aDepartments of Thoracic and Cardiovascular Surgery,bCardiology and Angiology, and cAnesthesiology and Operative Intensive Care Medicine, and the dInstitute of Clinical Chemistry and Laboratory Medicine, University Hospital Muenster, Muenster, Germany. Submitted April 28, 2004; revised July 8, 2004; accepted July 19, 2004. Reprint requests: Markus Rothenburger, MD, Department of Thoracic and Cardiovascular Surgery, University Hospital Muenster, Albert Schweitzer Strasse 33, D-48129 Muenster, Germany. Telephone: ⫹49-251-83-47401. Fax: ⫹49-251-83-48316. E-mail: [email protected] Copyright © 2004 by the International Society for Heart and Lung Transplantation. 1053-2498/04/$–see front matter. doi:10.1016/ j.healun.2004.07.006
CHF can be difficult to diagnose. Symptoms such as shortness of breath, fatigue, and peripheral edema can be nonspecific, and in mild cases, typical features of CHF may be absent.2–7 The classification according to the New York Heart Association (NYHA), although useful, is subjective and depends on the patients’ description of symptoms and the physicians’ experience. Different types of heart failure scores have been proposed and introduced for the clinical assessment of heart failure.8 All these scores offer good sensitivity and accuracy of heart failure severity, but require a wide spectrum of special and expensive examinations. Recent studies have increased the interest in the cardiac neurohormone B-type natriuretic peptide (BNP).1,2,9 BNP is specifically secreted from the cardiac ventricles as a response to ventricular volume expansion, pressure overload, and resultant increase in wall tension.10 –13 BNP is valuable in distinguishing dyspnea from pulmonary or cardiac causes, particularly when due to CHF. In patients with normal left ventricular 1189
1190
Rothenburger et al.
function assessed by echocardiography, BNP values are low.14 –17 A significant correlation between the NYHA functional classification or left ventricular ejection fraction and BNP has been reported.12,18,19 Cardiac transplantation is the only currently acknowledged definitive treatment for refractory heart failure. The selection criteria for heart transplant recipients include the risk factors associated with mortality.8 Selecting candidates for heart transplantation is sometimes difficult, because some variables have changed from absolute to relative contraindications for heart transplantation.20 In clinical decision making about patient candidacy for heart transplantation, a wide spectrum of special examinations are needed, frequently resulting in controversial discussions whether or not the patient is a candidate for heart transplantation. In this study, we investigated the power of NTproBNP levels in patients with different causes of severe heart failure as a selection criterion for cardiac transplant candidates in comparison to the Heart Failure Survival Score (HFSS) and the NYHA classification. Furthermore, we studied the impact of NT-proBNP as a predictive and prognostic marker in CHF patients. MATERIAL AND METHODS Patients Between January 2002 and January 2004, a total of 550 consecutive patients with heart failure due to left ventricular dysfunction, either as a result of coronary artery disease (CAD) or dilative cardiomyopathy (DCM), presenting to our Interdisciplinary Heart Failure Program were investigated. No exclusion criteria were defined. Patients in good clinical conditions (NYHA Class 1) served as reference group. The demographic data and patient characteristics are shown in Table 1. In all patients, NYHA classification, electrocardiogram (ECG), echocardiography, blood sampling (NT-proBNP, routine blood values), spiroergometry, and right heart catheterization were performed. The decision whether patients were potential candidates for heart transplantation were made by 2 cardiologists who were blinded to the results of NT-proBNP levels. For decision making, they used clinical condition (comorbidity); NYHA classification; invasive and noninvasive parameters such as echocardiography, ECG, coronary angiography, spiroergometry; and routine laboratory findings. Echocardiography Measurements of left ventricular dimensions were made from M-mode and 2-dimensional echocardiographic images in the parasternal long-axis view. Left ventricular volumes and ejection fractions were calculated by modification of Simpson’s method with 2 apical views.
The Journal of Heart and Lung Transplantation
Spiroergometry Patients underwent bicycle spiroergometry (Oxycon Pro; Jäger, Heidelberg, Germany) with 12-lead ECG monitoring in a half-sitting position to assess exercise capacity and maximum oxygen consumption as a marker of heart failure severity. The end point of exercise was defined as reaching the anaerobic threshold (RER ⬎1.0) or if severe dyspnea or angina pectoris occurred. The exercise capacity was assessed on the basis of maximum load levels expressed in watts ˙ O2max) and (Wmax), maximum peak oxygen uptake (V metabolic equivalents. The maximum oxygen con˙ O2max (ml/min/kg). sumption was calculated as V Measurement of NT-proBNP Blood samples were collected and NT-proBNP was measured by the ElectroChemiLumineszenz immunoassay (ECLIA) with the Elecsys system, 2010 Modular Analytics E170 (Roche Diagnostics, Mannheim, Germany), according to the manufacturer’s instructions. A biotinylated polyclonal antibody against NTproBNP and a polyclonal antibody coupled with a Ruthenium complex were used following the sandwich principle. Together with streptavidin, a biotin-streptavidin complex binds to the solid phase, which can be photometrically measured by magnetic interferences (so-called chemilumineszenz emission).
Table 1. Baseline Characteristics by Underlying Disease* Dilative Coronary artery Characteristic cardiomyopathy disease No. of patients 323 227 Age, y 55 ⫾ 11 57 ⫾ 8 Sex (M/F) 235/88 175/52 Height, cm 175 ⫾ 10 172 ⫾ 8 Weight, kg 80 ⫾ 17 80 ⫾ 14 Heart rate, bpm 67 ⫾ 14 69 ⫾ 11 NYHA class 2⫾1 3⫾1 LV-EF, % 32 ⫾ 13 31 ⫾ 11 FS, % 21 ⫾ 9 21 ⫾ 7 LVEDD, mm 68 ⫾ 11 66 ⫾ 8 LVESD, mm 54 ⫾ 13 52 ⫾ 9 Mean arterial pressure, 75 ⫾ 14 74 ⫾ 14 mm Hg Cardiac output, L/min 4.8 ⫾ 1.1 4.5 ⫾ 0.8 Cardiac index, L/min/m2 2.9 ⫾ 0.8 2.7 ⫾ 0.7 Serum sodium, mg/dl 138 ⫾ 3 139 ⫾ 2 Serum creatinine, mg/dl 1.5 ⫾ 0.8 1.3 ⫾ 0.4 V˙O2 max, ml/min/kg 15.9 ⫾ 4.2 14.9 ⫾ 3.5 Heart Failure Survival 8.75 ⫾ 1.16 8.63 ⫾ 0.98 Score
p Value
⬍0.01 NS NS NS NS NS ⬍0.01 ⬍0.01 NS 0.041 NS NS NS 0.004 NS
*LV-EF, left ventricular ejection fraction; FS, shortening fraction; LVEDD, left ventricular enddiastolic diameter; LVESD, left ventricular endsystolic diameter.
The Journal of Heart and Lung Transplantation
Right Heart Catheterization A Swan-Ganz catheter was placed via the jugular vein into the pulmonary artery. Hemodynamic parameters included pulmonary artery pressure, pulmonary capillary wedge pressure, cardiac output, cardiac index, systemic vascular resistance, and pulmonary vascular resistance. The arterial pressure was noninvasively measured by the following equation: mean arterial pressure ⫽ [systolic pressure ⫹ (2 ⫻ diastolic pressure)/3]. Definition of HFSS HFSS has been used to predict 1-year mortality in patients with CHF.8 The following parameters were included in the model: CAD (0 if absent, 1 if present); resting heart rate (bpm); left ventricular ejection fraction (from 0% to 100%); mean blood pressure (mm Hg); intraventricular conduction defect (0 if absent, 1 if present); peak oxygen consumption (ml/kg/min); serum sodium (mmol/liter), and invasive mean pulmonary capillary wedge pressure (mm Hg). The coefficients reported by Aaronson8 were used. Follow-up Patients admitted to our Interdisciplinary Heart Failure Program underwent routinely performed follow-up examinations for 3 to 6 months. The follow-up examinations included ECG, echocardiography, routine blood check, and assessment of NT-proBNP levels. Repeated spiroergometry and right heart catheterization were performed yearly. Statistical Analysis End points of our study were decision for cardiac transplantation, and for prognostic results, death of the patient during observation. Baseline characteristics are reported as mean ⫾ SD as appropriate. Univariate comparisons were made with 2 (with Yates correction for low expected frequencies) or 2-sample Student’s t-test as appropriate. The Pearson product moment correlation coefficient was used to measure linear correlations between variables. For correlation analysis, the log of NT-proBNP was used to normalize the distribution area. Correlations were compared by a z-test on normally transformed correlation coefficients. The relative abilities of NT-proBNP to predict the selection criteria for cardiac transplantation were assessed by receiver operating characteristic (ROC) analysis. AUC for each marker was compared by the method of Hanley and McNeil.21 Decision statistics were computed from 2 ⫻ 2 tables and reported as sensitivity, specificity, and positive and negative predictive values. Diagnostic accuracy was computed as the sum of the concordant cells divided by the sum of all cells in the 2 ⫻ 2 table. Agreement between clinical judgment and
Rothenburger et al.
1191
NT-proBNP was quantified by Cohen’s statistic. The positive likelihood ratio was taken as the slope of the ROC curve for the optimum cutpoint and was expressed as follows: (sensitivity/1 ⫺ specificity). Multivariate analysis was performed by the Cox binary logistic regression model. Statistical significance was assumed for a p value less than 0.05. RESULTS Baseline Characteristics A total of 550 patients were prospectively investigated. The demographics and baseline characteristics for the study sample according to the underlying heart disease are reported in Table 1. No differences between groups regarding weight, heart rate, ejection fraction, shortening fraction, mean arterial pressure, cardiac index, serum sodium, serum creatinine, and HFSS were observed. Patients with DCM were significantly taller (p ⬍ 0.01). Their ventricular diameters left ventricular enddiastolic diameter (LVEDD) and left ventricular endsystolic diameter (LVESD) were greater (p ⬍ 0.01), the cardiac output was higher (p ⫽ 0.041), and the maximum oxygen consumption was greater when compared with the ischemic patients (p ⫽ 0.004; Table 1). NT-proBNP and NYHA Class The baseline characteristics of patients grouped according to NYHA functional class are presented in Table 2. NT-proBNP levels were significantly higher with increasing NYHA class (p ⫽ 0.0024; Figure 1). Patients in NYHA Class 1 served as the reference group. In these 51 patients, the mean level of NT-proBNP was 50 ⫾ 13 ng/ml. The levels increased from NYHA Class 2 (150 ⫾ 56 ng/ml) to NYHA Class 3 (1800 ⫾ 452 ng/ml) and NYHA Class 4 (3800 ⫾ 499 ng/ml). NT-proBNP and HFSS The baseline characteristics by HFSS risk group are reported in Table 3. NT-proBNP levels were significantly higher in the HFSS high-risk group (p ⫽ 0.013; Figure 2). Low-risk patients had low NT-proBNP levels of (249 ⫾ 64 ng/ml), whereas the medium-risk patients (950 ⫾ 97 ng/ml) and high-risk patients (2689 ⫾ 552 ng/ml) showed significantly higher values. NT-proBNP and Selection of Transplant Candidates NT-proBNP levels of 254 patients assigned to heart transplantation were significantly higher than in 296 patients who were not assigned (2293 pg/ml vs 493 pg/ml; p ⬍ 0.001). In patients with DCM, the NT-proBNP levels were significantly higher in the group considered for heart transplantation (2789 ⫾ 467 ng/ml vs 280 ⫾ 123 ng/ml; p ⬍ 0.001; Figure 3). Patients with CAD also showed higher NT-proBNP levels in the group assigned for heart
1192
Rothenburger et al.
The Journal of Heart and Lung Transplantation
Table 2. Patient Characteristics by NYHA Functional Class* Characteristic No. of patients Age, y Sex (M/F) QRS ⬎0.12, % LV-EF, % LVEDD, mm LVESD, mm FS, % Mean arterial pressure, mm Hg Cardiac output, L/min Cardiac index, L/min/m2 Serum sodium, mg/dl Serum creatinine, mg/dl V˙ O2 max., ml/min/kg Heart Failure Survival Score
NYHA class 1 51 46 ⫾ 10 26/25 21,1 38 ⫾ 13 65 ⫾ 10 50 ⫾ 10 24 ⫾ 6 79 ⫾ 11 5.7 ⫾ 0.6 3.9 ⫾ 0.6 138 ⫾ 1 1.0 ⫾ 0.2 21.4 ⫾ 4.7 9.72 ⫾ 1
NYHA class 2 223 57 ⫾ 9 191/32 65.2 33 ⫾ 11 68 ⫾ 10 54 ⫾ 12 21 ⫾ 8 75 ⫾ 13 5.0 ⫾ 1 3 ⫾ 0.7 139 ⫾ 3 1.3 ⫾ 0.7 16.8 ⫾ 3.1 8.96 ⫾ 0.92
NYHA class 3 262 53 ⫾ 13 170/92 77.5 25 ⫾ 8 67 ⫾ 10 54 ⫾ 11 18 ⫾ 6 66 ⫾ 13 4.3 ⫾ 1 2.4 ⫾ 0.6 138 ⫾ 3 1.4 ⫾ 0.5 13.9 ⫾ 3.3 8.03 ⫾ 0.76
NYHA class 4 14 54 ⫾ 10 10/4 100 23 ⫾ 12 69 ⫾ 12 60 ⫾ 15 14 ⫾ 7 60 ⫾ 4 3.7 ⫾ 0.3 2.0 ⫾ 0.2 136 ⫾ 5 2.4 ⫾ 1.2 11.1 ⫾ 2.3 7.08 ⫾ 0.86
p Value ⬍0.001 ⬍0.001 ⬍0.01 NS NS ⬍0.001 ⬍0.01 ⬍0.001 ⬍0.001 ⬍0.02 ⬍0.001 ⬍0.001 ⬍0.001
*LV-EF, left ventricular ejection fraction; FS, shortening fraction; LVEDD, left ventricular enddiastolic diameter; LVESD, left ventricular endsystolic diameter.
transplantation (1800 ⫾ 430 ng/ml vs 156 ⫾ 68 ng/ml; p ⬍ 0.01). Results of the univariate and multivariate analysis of different parameters are shown in Table 4. Receiver Operating Characteristic Curves In decision making for or against heart transplantation, arbitrary cutoff points for NT-proBNP were used. For patients with DCM, an arbitrary cutoff point of 2000 pg/ml was defined. At this cutoff point, a positive predictive value of 0.87, a negative predictive value of 0.89, a sensitivity of 0.84, and a specificity of 0.88 were found, resulting in an odds ratio of 7.65 (p ⬍ 0.001). Compared with NYHA and HFSS, NT-proBNP the results
were significantly better (p ⬍ 0.001). For patients with CAD, an arbitrary cutoff point of 1000 pg/ml was defined. At this cutoff point, a positive predictive value of 0.96, a negative predictive value of 0.87, a sensitivity of 0.84, and a specificity of 0.96 were found, resulting in an odds ratio of 5.14 (p ⬍ 0.001). Compared with NYHA and HFSS, NT-proBNP the results were significantly better (p ⬍ 0.001). By use of an arbitrary cutoff point of 2000 mg/ml, regarding the decision for heart transplant candidacy showed, the use of the HFSS only revealed an AUC of 0.847 ⫾ 0.017 (range 0.814 – 0.880; Figure 4). Making decisions that used the NYHA classification only showed an AUC of 0.859 ⫾ 0.017 (range 0.825– 0.892), and the use of NT-proBNP 0.962 ⫾ 0.008 (range 0.945– 0.978; p ⬍ 0.001 for pairwise comparisons; Figure 4). The combined use of all 3 parameters improved the AUC to 0.984 ⫾ 0.012 (range 0.944 – 0.989) and therefore the significance of the decision for heart transplant candidacy. NT-proBNP as a Predictive Marker for Treatment Consequences
Figure 1. NT-proBNP in correlation to New York Heart Association (NYHA) functional classification; *p ⬍ 0.001 significances NYHA Class 3 and 4 vs other groups. BNP, B-type natriuretic peptide.
The mean NT-proBNP levels of all patients NYHA Class 2 or more (n ⫽ 276, 50.2%) who were admitted for the first time to our interdisciplinary heart failure office were 2500 ⫾ 560 pg/ml. As a consequence of optimized drug treatment among NYHA Class 3 and 4 patients, NT-proBNP levels decreased (3 ⫾ 1 month) in 187 patients (67.7%; NT-proBNP 876 ⫾ 279 pg/ml vs 2670 ⫾ 564 pg/ml; p ⫽ 0.023). NT-proBNP remained elevated or increased in 89 patients (32.3%; NT-proBNP 3217 ⫾ 829 pg/ml vs 2760 ⫾ 564 pg/ml; p ⫽ 0.03). Of these, 10 patients were treated immediately with a left ventricular assist device (LVAD).
The Journal of Heart and Lung Transplantation
Rothenburger et al.
1193
Table 3. Patient Characteristics by HFSS Risk Group* HFSS Characteristic No. of patients Age, years Sex (M/F) QRS ⬎ 0.12, % NYHA class LV-EF, % LVEDD, mm LVESD, mm FS, % Mean arterial pressure, mm Hg Cardiac output, L/min Cardiac index, L/min/m2 Serum sodium, mg/dl Serum creatinine, mg/dl V˙ O2 max., ml/min/kg Heart Failure Survival Score
High risk 58 58 ⫾ 10 42/16 91.6 3⫾0 22 ⫾ 4 72 ⫾ 9 59 ⫾ 9 15 ⫾ 4 60 ⫾ 7 3.7 ⫾ 1.2 1.9 ⫾ 0.6 137 ⫾ 3 1.7 ⫾ 0.8 11.9 ⫾ 3 6.95 ⫾ 0.27
Medium risk 166 54 ⫾ 9 128/38 85.7 3⫾1 22 ⫾ 4 70 ⫾ 7 58 ⫾ 8 17 ⫾ 5 62 ⫾ 7 4.4 ⫾ 1 2.5 ⫾ 0.7 138 ⫾ 3 1.4 ⫾ 0.7 14.5 ⫾ 3.7 7.76 ⫾ 0.28
Low risk 326 54 ⫾ 13 224/102 26.2 2⫾1 35 ⫾ 11 65 ⫾ 10 52 ⫾ 12 22 ⫾ 8 76 ⫾ 14 5.0 ⫾ 0.9 3.0 ⫾ 0.7 139 ⫾ 2 1.3 ⫾ 0.5 16.9 ⫾ 3.8 9.16 ⫾ 0.81
p Value NS ⬍0.02 ⬍0.01 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.01 ⬍0.001 ⬍0.001 NS NS ⬍0.001 ⬍0.001
*LV-EF, left ventricular ejection fraction; FS, shortening fraction; LVEDD, left ventricular enddiastolic diameter; LVESD, left ventricular endsystolic diameter.
Two patients underwent urgent heart transplantation. An implantable cardioverter system (ICD) was implanted in 39 patients. NT-proBNP as a Prognostic Marker in CHF Patients The NT-proBNP levels above 5000 pg/ml were in 58 patients (21% of NYHA Class 3 and 4 patients). The mortality rate among NYHA Class 3 and 4 patients was 10.2% (n ⫽ 28). The survivors in this group (30 of 58 patients) were treated with either LVAD support (n ⫽
Figure 2. NT-proBNP and correlation to risk groups of Heart Failure Survival Score; significance high risk group vs other groups *p ⬍ 0.01. BNP, B-type natriuretic peptide.
10) or urgent heart transplantation (n ⫽ 2). In 18 patients, an ICD system implantation were performed. During the follow-up period of 2 years, 28 of 276 NYHA Class 3 and 4 patients (10.2%) died while on the waiting list for heart transplantation (n ⫽ 9) or during evaluation for heart transplantation (n ⫽ 19). The reasons of death were heart failure (n ⫽ 17), severe arrhythmias (n ⫽ 8), and noncardiac causes (n ⫽ 3). The mean NT-proBNP
Figure 3. NT-proBNP levels in patients with chronic heart failure assigned to cardiac transplantation or not; *p ⬍ 0.001 significance of decision differences for each group. BNP, B-type natriuretic peptide; DCM, dilative cardiomyopathy; CAD, coronary artery disease.
1194
Rothenburger et al.
The Journal of Heart and Lung Transplantation
Table 4. Factors for Decision Making for Cardiac Transplantation* Variable Age (⬎60 years) Weight (⬎85 kg; ⬍40 kg) Height (⬎180 cm; ⬍120 cm) CAD HFSS (⬍ 8.09) NYHA class (3/4) NT-proBNP (⬎1.000 ng/ml) V˙ O2max (⬍12 ml/min/kg) Creatinine (⬍1.5 mg/dl) Sodium (⬍133 mmol/l) Heart rate (⬎100 bpm) QRS (⬎0.12ms) CO (⬍4.0 L/min) CI (⬍2.0 L/min/m2) EF (⬍30%) FS (⬍25%) LVEDD (⬎ 6 cm) LVESD (⬎4 cm)
p Value (univariate) 0.237 0.017 0.314 0.012 ⬍0.001 ⬍0.001 ⬍0.001 0.042 0.404 0.321 0.052 0.023 0.962 0.051 0.031 0.364 0.779 0.529
OR (CI 95%) 0.48 (0.28–7.22) 3.47 (1.94–8.23) 0.66 (0.12–8.49) 2.4 (1.01–7.43) 7.42 (4.5–9.87) 8.48 (3.9–11.9) 10.6 (3.7–14.5) 2.07 (1.26–5.11) 1.54 (0.4–7.62) 1.29 (0.07–8.23) 2.68 (1.2–7.53) 2.47 (1.26–5.11) 1.08 (0.4–9.32) 1.45 (0.62–5.8) 2.21 (1.03–7.67) 1.32 (0.79–3.45) 1.08 (0.7–6.31) 1.12 (0.4–7.21)
p Value (multivariate) 0.28 0.008 0.26 0.039 0.02 0.01 0.01 0.58 0.12 0.19 0.15 0.46 0.90 0.56 0.02 0.41 0.80 0.48
*CI, confidence interval; OR, odds ratio; CAD, coronary artery disease; CO, cardiac output; CI, cardiac index; EF, ejection fraction; FS, shortening fraction; LVEDD, left ventricular enddiastolic diameter; LVESD, left ventricular endsystolic diameter.
level in these patients was 5423 ⫾ 423 pg/ml. The mortality rate significantly correlated with NT-proBNP levels (r ⫽ 0.71; p ⬍ 0.02). Patients with NT-proBNP levels below 1000 pg/ml at admission had a yearly mortality rate of 0.7%; for levels of 1000 to 4999 pg/ml, the mortality rate was 1.6%; and for levels above 5000 pg/ml, the mortality rate was 28.4% (p ⫽ 0.014).
DISCUSSION The poor prognosis associated with advanced heart failure has been demonstrated by Gardner and coworkers.22 They described a 1-year mortality of approximately 15% among patients suffering from advanced heart failure. The recent COPERNICUS Study has provided comparable 1-year mortality rates of 18.5% in the placebo-treated group and 11.4% in the carvediloltreated group.23,24 The high mortality rates of end-stage heart failure make it desirable that parameters be found that help identify patients at high risk of death or adverse outcome. These patients can then be considered for cardiac transplantation or other surgical treatment, such as insertion of LVADs. The present study provides a large single-center comparison of prognostic assessment by use of HFSS in patients with heart failure due to different causes. Our study demonstrates that the measurement of NTproBNP concentrations may serve as an independent marker of mortality, and it provides important information for clinical decision making in heart transplant or LVAD candidates. The study further provides the predictive and prognostic impact of NT-proBNP levels on treatment options and mortality of CHF patients. NT-proBNP and Etiology
Figure 4. Receiver operating characteristic (ROC) curves for the sensitivity and specificity of NT-proBNP, Heart Failure Survival Score (HFSS), and New York Heart Association (NYHA) class and assignment to heart transplantation; (p ⬍ 0.001 for all pairwise comparisons of area under ROC curve). BNP, B-type natriuretic peptide.
The levels of NT-proBNP in patients with DCM were significantly higher than in patients with left ventricular dysfunction due to coronary artery disease. There was a strong correlation between NYHA class and HFSS to NT-proBNP. Potential reasons why increased left ven-
The Journal of Heart and Lung Transplantation
tricular diameters and ventricular wall tension results in significantly higher NT-proBNP levels in patients with DCM have been discussed.13,22 Consensus exists that increasing levels of NT-proBNP strongly correlate with both increased ventricular chamber diameters and increased ventricular wall tension. Recent reports indicate that cardiac neurohormone levels reflect left ventricular function. They add useful information beyond clinical features, including results of cardiac imaging, in staging and predicting outcome of patients with CHF due to dilated cardiomyopathy or after myocardial infarction.9,25,26 As a putative measure of the hemodynamic consequences of heart failure in myocardial ischemia or after myocardial infarction, BNP has been shown to be useful for risk assessment among patients with non-ST elevation acute coronary syndrome.1,26 Fewer data are available in advanced left ventricular dysfunction due to coronary artery disease. The impact of NT-proBNP in patients after multiple myocardial infarctions without options for revascularization, even without symptoms such as angina pectoris, have not been elucidated. However, NT-proBNP is markedly lower in left ventricular dysfunction due to coronary artery disease compared with DCM, despite similar ejection fraction or NYHA class. The results of our study demonstrate a close correlation to NYHA classification and HFSS, and confirm the prognostic impact of NT-proBNP in patients with advanced heart failure. NT-proBNP, and HFSS and NYHA Functional Class Heart failure can be difficult to diagnose accurately because the signs and symptoms of this disorder are neither sensitive nor specific. These limitations are especially relevant when symptoms are mild, or when patients are elderly or have comorbid disorders that mimic heart failure, such as pulmonary disease or obesity. Studies have shown that BNP may help to distinguish patients suffering from dyspnea due to heart failure or pulmonary reasons.2,5–7,17 By use of a BNP threshold of more than 100 pg/ml to diagnose heart failure, BNP did better than all other clinical variables and the clinical judgment of the emergency room physician. It contributed explanatory power to a multivariable model that incorporated diagnostic variables from the patient’s history, examination, and chest x-ray.2,5–7 Asymptomatic left ventricular dysfunction is at least as common as symptomatic heart failure. A simple screening test might help identify patients at risk of developing heart failure who would benefit from treatment that prevents progression to heart failure, including angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and -blockers. Such screening could be targeted to patients at high risk for left
Rothenburger et al.
1195
ventricular dysfunction, such as those with diabetes, recent myocardial infarction, or end-stage renal failure.27–29 NT-proBNP and Decision Making in Selection of Heart Transplant Candidates Beyond its diagnostic and prognostic value, NT-proBNP can also help physicians in clinical decision making, particularly with regard to the question of transplant candidacy in patients with advanced heart failure. The premise of this approach is that decisions to initiate or titrate pharmacological treatments, or to use a more invasive strategy such as cardiac transplantation, might be based not only on symptoms and physical examination findings, but also on plasma levels of NT-proBNP. Thus, NT-proBNP could ultimately prove useful in helping physicians to select the appropriate drugs and drug doses.30 Further data are needed to demonstrate the potential value of NT-proBNP to assess the need for more invasive strategies such as implantable defibrillators, ventricular assist devices, or cardiac transplantation. The present study shows a good correlation of the levels of the neurohormone NT-proBNP and the consideration for cardiac transplantation. On the basis of clinical findings, high levels of NT-proBNP were strongly associated with a high risk of fatal outcome. On the basis of these findings, we suggest an arbitrary cutoff point of 2000 pg/ml for patients with DCM and of 1000 pg/ml for patients with severe left ventricular dysfunction due to coronary artery disease. More than 90% of all patients above these cutoff points of NTproBNP were clinically considered for heart transplant candidacy when examined by blinded cardiologists. Because these values showed a high sensitivity and specificity in decision making in heart transplant candidates, noninvasive NT-proBNP may help unravel the complex problem of whom to refer for cardiac transplantation. NT-proBNP as a Predictive Marker for Treatment Consequences Decision making for different treatment options might sometimes be difficult in CHF patients. In the present study, NT-proBNP provides a useful parameter in controlling treatment effects. Patients who initiate drug treatment could be followed up regarding treatment success, which results in a decrease of NT-proBNP levels. Although an improvement in NYHA classification and HFSS provides similar informations, the first one is subjective, and the second one requires many examinations before a decision is possible. Patients without any changes or increases in NT-proBNP due to drug treatment should be undergo either LVAD implantation or urgent heart transplantation.
1196
Rothenburger et al.
NT-proBNP as a Prognostic Marker In this study, NT-proBNP has been detected as an prognostic marker in CHF patients. Levels above 5000 pg/ml were associated with a markedly increased mortality rate. The intervention rate in these patients was significantly higher; most of them underwent LVAD implantation, urgent heart transplantation, or implantation of an ICD. Tsutamoto and coworkers31 reported that natriuretic peptides offer a new option in evaluating prognosis in patients with clinical heart failure. In contrast, Berger et al32 showed that big endothelin seems to be a better marker to estimate the prognosis of CHF than neuroendocrine hormones within the first year. Selvais et al33 reported that endothelin was superior to BNP in predicting 3-year mortality in a population of 109 patients with symptomatic heart failure. Nevertheless, our data provide NT-proBNP levels in CHF patients that have not been shown in terms of the prognostic value of this marker in previous studies. In our study, NT-proBNP was found to be highly associated with prognosis in CHF patients. Conclusions NT-proBNP was found to be a helpful marker in selection of heart transplant candidates. Furthermore, NTproBNP provides an effective parameter of treatment control in CHF patients. It is useful in decision making for further treatment options (medically, LVAD implantation or urgent heart transplantation) in CHF patients. The prognostic value of NT-proBNP may be helpful in finding patients for require immediate hospitalization for heart failure. REFERENCES 1. DeLemos JA, Morrow DA, Bentley JH, et al. The prognostic value of B-type natriuretic peptide in patients with acute coronary syndromes. N Engl J Med 2001;345:1014 – 21. 2. 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. 3. McCullough PA, Nowak RM, McCord J, et al. B-type natriuretic peptide and clinical judgement in emergency diagnosis of heart failure. Circulation 2002;106:416 –22. 4. Mair J, Hammerer-Lercher A, Puschendorf B. The impact of cardiac natriuretic peptide determination on the diagnosis and management of heart failure. Clin Chem Lab Med 2001;39:571– 88. 5. Vesely DL. Natriuretic peptides and acute renal failure. Am J Physiol Renal Physiol 2003;285:F167–77. 6. Maisel AS, McCord J, Nowak RM, et al. Bedside B-type natriuretic peptide in the emergency diagnosis of heart failure with reduced or preserved ejection fraction. J Am Coll Cardiol 2003;41:2010 –7.
The Journal of Heart and Lung Transplantation
7. Morrison LK, Harrison A, Krishnaswamy P, et al. Utility of a rapid B-natriuretic peptide assay in differentiating congestive heart failure from lung disease in patients presenting with dyspnoea. J Am Coll Cardiol 2002;39:202–9. 8. Aaronson KD, Schartz JS, Chen TM, et al. Development and prospective validation of a clinical index to predict survival in ambulatory patients referred for cardiac transplant evaluation. Circulation 1997;95:2660 –7. 9. Richards AM, Nicholls MG, Yandle TG, et al. Neuroendocrine prediction of left ventricular function and heart failure after acute myocardial infarction. Heart 1999;81: 114 –20. 10. Schrier RW, Abraham WT. Hormones and hemodynamics in heart failure. N Engl J Med 1999;341:577– 85. 11. Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med 1998;339:321– 8. 12. De Lemos JA, McGuire DK, Drazner MH. B-type natriuretic peptide in cardiovascular disease. Lancet 2003;362: 316 –22. 13. Ruskoaho H. Cardiac hormones as diagnostic tools in heart failure. Endocr Rev 2003;24:341–56. 14. Jourdain P, Funck F, Bellorini M, et al. Bedside B-type natriuretic peptide and functional capacity in chronic heart failure. Eur J Heart Failure 2003;5:155– 60. 15. Koch A, Singer H. Normal values of B type natriuretic peptide in infants, children, and adolescents. Heart 2003; 89:875– 8. 16. Raymond I, Groenning BA, Hildebrandt PR, et al. The influence of age, sex, and other variables on the plasma level of N-terminal pro brain natriuretic peptide in a large sample ot the general population. Heart 2003;89:745–51. 17. Goetze JP, Videbaek R, Boesgaard S, et al. Pro-brain natriuretic peptide as marker of cardiovascular or pulmonary causes of dyspnea in patients with terminal parenchymal lung disease. J Heart Lung Transplant 2004;23: 80 –7. 18. Richards AM, Nicholls MG, Espiner EA, et al. B-type natriuretic peptides and ejection fraction for prognosis after myocardial infarction. Circulation 2003;107:2786 – 92. 19. Wieczorek SJ, Wu AHB, Christenson R, et al. A rapid B-type natriuretic peptide assay accurately diagnoses left ventricular dysfunction and heart failure: a multicenter evaluation. Am Heart J 2002;144:834 –9. 20. Cimato TR, Jessup M. Recipient selection in cardiac transplantation: contraindications and risk factors for mortality. J Heart Lung Transplant 2002;21:1161–73. 21. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143:29 –36. 22. Gardner RS, Özalp F, Murday AJ, et al. N-terminal probrain natriuretic peptide. A new gold standard in predicting mortality in patients with advanced heart failure. Eur Heart J 2003;24:1735– 43. 23. Packer M, Coats AJ, Fowler MB, et al. Effect of carvedilol on survival in severe chronic heart failure. N Engl J Med 2001;344:1651– 8. 24. Packer M, Fowler MB, Roecker EB, et al. Effect of carvedilol on the morbidity of patients with severe chronic heart failure: results of the Carvedilol Prospective
The Journal of Heart and Lung Transplantation
25.
26.
27.
28.
29.
Randomized Cumulative Survival (COPERNICUS) study. Circulation 2002;106:2194 –9. Fisher C, Berry C, Blue L, et al. N-terminal pro B type natriuretic peptide, but not the new putative cardiac hormone relaxin, predicts prognosis in patients with chronic heart failure. Heart 2003;89:879 – 81. Goetze JP, Christoffersen C, Perko M, et al. Increased cardiac BNP expression associated with myocardial ischemia. FASEB J 2003;17:U24 –33. Struthers AD, Morris AD. Screening for and treating left-ventricular abnormalities in diabetes mellitus: a new way of reducing cardiac death. Lancet 2002;359:1430 –2. Motwani JG, McAlpine H, Kennedy N, et al. Plasma brain natriuretic peptide as an indicator for angiotensin-converting enzyme inhibition after myocardial infarction. Lancet 1993;341:1109 –13. Mallamaci F, Zoccali C, Tripepi G, et al. Diagnostic potential of cardiac natriuretic peptides in dialysis patients. Kidney Int 2001;59:1559 – 66.
Rothenburger et al.
1197
30. Richards AM, Doughty R, Nicholls MG, et al. Plasma N-terminal pro-brain natriuretic peptide and adrenomedullin: prognostic utility and prediction of benefit from carvedilol in chronic ischemic left ventricular dysfunction. J Am Coll Cardiol 2001;37:1781–7. 31. Tsutamoto T, Wada A, Maeda K, et al. Attenuation of compensation of endogenous cardiac natriuretic peptide system in chronic heart failure: prognostic role of plasma brain natriuretic peptide concentration in patients with chronic symptomatic left ventricular dysfunction. Circulation 1997;96:509 –16. 32. Berger R, Strecker K, Huelsmann M, et al. Prognostic power of neurohormonal parameters in chronic heart failure depends on clinical stage and observation period. J Heart Lung Transplant 2003;22:1037– 45. 33. Selvais PL, Robert A, Ahn S, et al. Direct comparison between endothelin-1, N-terminal proatrial natriuretic factor, and brain natriuretic peptide as prognostic markers of survival in congestive heart failure.
Aminoterminal pro Type B Natriuretic Peptide as a Predictive and Prognostic Marker in Patients With Chronic Heart Failure Markus Rothenburger, MD,a Thomas Wichter, MD,b Christof Schmid, MD,a Jörg Stypmann, MD,b Tonny D. T. Tjan, MD,a Elmar Berendes, MD,c Christian Etz, MD,a Aurélien Pioux, MD,d Andreas Löher, MD,a Frauke Wenzelburger, MD,a Gabriele Drees, MD,a Andreas Hoffmeier, MD,a Günter Breithardt, MD,b and Hans Heinrich Scheld, MDa Background: B-type natriuretic peptide (BNP) is released from the cardiac ventricles in response to increased wall tension. We studied the relation of NT-proBNP to Heart Failure Survival Score (HFSS) and New York Heart Association (NYHA) class in patients with chronic heart failure (CHF). We also studied the impact for recipient selection for cardiac transplant and assessed it as a predictive and prognostic marker of CHF. Methods: A total of 550 patients with dilative cardiomyopathy (n ⫽ 323), and coronary artery disease (n ⫽ 227) were prospectively examined. All patients underwent spiroergometry, echocardiography, right heart catheterization, and electrocardiogram. Routine blood levels and NT-proBNP were measured. The clinical selection for cardiac transplant candidates was adjudicated by 2 independent cardiologists who were blinded to the results of NT-proBNP assays. Clinical outcome and predictive power of NT-proBNP were analyzed. Results: NT-proBNP levels in patients clinically considered for cardiac transplantation were significantly higher (2293 ng/ml vs 493 ng/ml; p ⬍ 0.001). The receiver operating characteristic (ROC) analysis regarding transplant candidacy showed an area under the ROC curve (AUC) of 0.84 ⫾ 0.01 for HFSS, 0.86 ⫾ 0.001 for NYHA, and 0.96 ⫾ 0.01 for NT-proBNP. Patients with increasing NT-proBNP levels or remaining elevated levels despite adequate heart insufficiency treatment were maintained with left ventricular assist device implantation (n ⫽ 10) or urgent heart transplantation (n ⫽ 2). Patients with NT-proBNP levels above 5000 pg/ml had a mortality rate of 28.4% per year. Twenty-eight patients died during the observation period; all these patients were within NYHA Classes 3 and 4 (NT-proBNP 5423 ⫾ 423 ng/ml). Conclusions: NT-proBNP discriminates patients at high likelihood of being a candidate for transplantation and provides prognostic informations in patients with CHF. NT-proBNP levels above 5000 pg/ml at admission were associated with death, and these levels markedly discriminated candidates for left ventricular assist devices or urgent transplantation. J Heart Lung Transplant.
An increasing number of patients with chronic heart failure (CHF) may have significant morbidity and mortality, with 65% of patients dying within 5 years.1 The health care expenditure for this disease is substantial, accounting for 1% to 2% of total health care costs, most of which is the result of repeated hospitalizations.2
From the aDepartments of Thoracic and Cardiovascular Surgery,bCardiology and Angiology, and cAnesthesiology and Operative Intensive Care Medicine, and the dInstitute of Clinical Chemistry and Laboratory Medicine, University Hospital Muenster, Muenster, Germany. Submitted April 28, 2004; revised July 8, 2004; accepted July 19, 2004. Reprint requests: Markus Rothenburger, MD, Department of Thoracic and Cardiovascular Surgery, University Hospital Muenster, Albert Schweitzer Strasse 33, D-48129 Muenster, Germany. Telephone: ⫹49-251-83-47401. Fax: ⫹49-251-83-48316. E-mail: [email protected] Copyright © 2004 by the International Society for Heart and Lung Transplantation. 1053-2498/04/$–see front matter. doi:10.1016/ j.healun.2004.07.006
CHF can be difficult to diagnose. Symptoms such as shortness of breath, fatigue, and peripheral edema can be nonspecific, and in mild cases, typical features of CHF may be absent.2–7 The classification according to the New York Heart Association (NYHA), although useful, is subjective and depends on the patients’ description of symptoms and the physicians’ experience. Different types of heart failure scores have been proposed and introduced for the clinical assessment of heart failure.8 All these scores offer good sensitivity and accuracy of heart failure severity, but require a wide spectrum of special and expensive examinations. Recent studies have increased the interest in the cardiac neurohormone B-type natriuretic peptide (BNP).1,2,9 BNP is specifically secreted from the cardiac ventricles as a response to ventricular volume expansion, pressure overload, and resultant increase in wall tension.10 –13 BNP is valuable in distinguishing dyspnea from pulmonary or cardiac causes, particularly when due to CHF. In patients with normal left ventricular 1189
1190
Rothenburger et al.
function assessed by echocardiography, BNP values are low.14 –17 A significant correlation between the NYHA functional classification or left ventricular ejection fraction and BNP has been reported.12,18,19 Cardiac transplantation is the only currently acknowledged definitive treatment for refractory heart failure. The selection criteria for heart transplant recipients include the risk factors associated with mortality.8 Selecting candidates for heart transplantation is sometimes difficult, because some variables have changed from absolute to relative contraindications for heart transplantation.20 In clinical decision making about patient candidacy for heart transplantation, a wide spectrum of special examinations are needed, frequently resulting in controversial discussions whether or not the patient is a candidate for heart transplantation. In this study, we investigated the power of NTproBNP levels in patients with different causes of severe heart failure as a selection criterion for cardiac transplant candidates in comparison to the Heart Failure Survival Score (HFSS) and the NYHA classification. Furthermore, we studied the impact of NT-proBNP as a predictive and prognostic marker in CHF patients. MATERIAL AND METHODS Patients Between January 2002 and January 2004, a total of 550 consecutive patients with heart failure due to left ventricular dysfunction, either as a result of coronary artery disease (CAD) or dilative cardiomyopathy (DCM), presenting to our Interdisciplinary Heart Failure Program were investigated. No exclusion criteria were defined. Patients in good clinical conditions (NYHA Class 1) served as reference group. The demographic data and patient characteristics are shown in Table 1. In all patients, NYHA classification, electrocardiogram (ECG), echocardiography, blood sampling (NT-proBNP, routine blood values), spiroergometry, and right heart catheterization were performed. The decision whether patients were potential candidates for heart transplantation were made by 2 cardiologists who were blinded to the results of NT-proBNP levels. For decision making, they used clinical condition (comorbidity); NYHA classification; invasive and noninvasive parameters such as echocardiography, ECG, coronary angiography, spiroergometry; and routine laboratory findings. Echocardiography Measurements of left ventricular dimensions were made from M-mode and 2-dimensional echocardiographic images in the parasternal long-axis view. Left ventricular volumes and ejection fractions were calculated by modification of Simpson’s method with 2 apical views.
The Journal of Heart and Lung Transplantation
Spiroergometry Patients underwent bicycle spiroergometry (Oxycon Pro; Jäger, Heidelberg, Germany) with 12-lead ECG monitoring in a half-sitting position to assess exercise capacity and maximum oxygen consumption as a marker of heart failure severity. The end point of exercise was defined as reaching the anaerobic threshold (RER ⬎1.0) or if severe dyspnea or angina pectoris occurred. The exercise capacity was assessed on the basis of maximum load levels expressed in watts ˙ O2max) and (Wmax), maximum peak oxygen uptake (V metabolic equivalents. The maximum oxygen con˙ O2max (ml/min/kg). sumption was calculated as V Measurement of NT-proBNP Blood samples were collected and NT-proBNP was measured by the ElectroChemiLumineszenz immunoassay (ECLIA) with the Elecsys system, 2010 Modular Analytics E170 (Roche Diagnostics, Mannheim, Germany), according to the manufacturer’s instructions. A biotinylated polyclonal antibody against NTproBNP and a polyclonal antibody coupled with a Ruthenium complex were used following the sandwich principle. Together with streptavidin, a biotin-streptavidin complex binds to the solid phase, which can be photometrically measured by magnetic interferences (so-called chemilumineszenz emission).
Table 1. Baseline Characteristics by Underlying Disease* Dilative Coronary artery Characteristic cardiomyopathy disease No. of patients 323 227 Age, y 55 ⫾ 11 57 ⫾ 8 Sex (M/F) 235/88 175/52 Height, cm 175 ⫾ 10 172 ⫾ 8 Weight, kg 80 ⫾ 17 80 ⫾ 14 Heart rate, bpm 67 ⫾ 14 69 ⫾ 11 NYHA class 2⫾1 3⫾1 LV-EF, % 32 ⫾ 13 31 ⫾ 11 FS, % 21 ⫾ 9 21 ⫾ 7 LVEDD, mm 68 ⫾ 11 66 ⫾ 8 LVESD, mm 54 ⫾ 13 52 ⫾ 9 Mean arterial pressure, 75 ⫾ 14 74 ⫾ 14 mm Hg Cardiac output, L/min 4.8 ⫾ 1.1 4.5 ⫾ 0.8 Cardiac index, L/min/m2 2.9 ⫾ 0.8 2.7 ⫾ 0.7 Serum sodium, mg/dl 138 ⫾ 3 139 ⫾ 2 Serum creatinine, mg/dl 1.5 ⫾ 0.8 1.3 ⫾ 0.4 V˙O2 max, ml/min/kg 15.9 ⫾ 4.2 14.9 ⫾ 3.5 Heart Failure Survival 8.75 ⫾ 1.16 8.63 ⫾ 0.98 Score
p Value
⬍0.01 NS NS NS NS NS ⬍0.01 ⬍0.01 NS 0.041 NS NS NS 0.004 NS
*LV-EF, left ventricular ejection fraction; FS, shortening fraction; LVEDD, left ventricular enddiastolic diameter; LVESD, left ventricular endsystolic diameter.
The Journal of Heart and Lung Transplantation
Right Heart Catheterization A Swan-Ganz catheter was placed via the jugular vein into the pulmonary artery. Hemodynamic parameters included pulmonary artery pressure, pulmonary capillary wedge pressure, cardiac output, cardiac index, systemic vascular resistance, and pulmonary vascular resistance. The arterial pressure was noninvasively measured by the following equation: mean arterial pressure ⫽ [systolic pressure ⫹ (2 ⫻ diastolic pressure)/3]. Definition of HFSS HFSS has been used to predict 1-year mortality in patients with CHF.8 The following parameters were included in the model: CAD (0 if absent, 1 if present); resting heart rate (bpm); left ventricular ejection fraction (from 0% to 100%); mean blood pressure (mm Hg); intraventricular conduction defect (0 if absent, 1 if present); peak oxygen consumption (ml/kg/min); serum sodium (mmol/liter), and invasive mean pulmonary capillary wedge pressure (mm Hg). The coefficients reported by Aaronson8 were used. Follow-up Patients admitted to our Interdisciplinary Heart Failure Program underwent routinely performed follow-up examinations for 3 to 6 months. The follow-up examinations included ECG, echocardiography, routine blood check, and assessment of NT-proBNP levels. Repeated spiroergometry and right heart catheterization were performed yearly. Statistical Analysis End points of our study were decision for cardiac transplantation, and for prognostic results, death of the patient during observation. Baseline characteristics are reported as mean ⫾ SD as appropriate. Univariate comparisons were made with 2 (with Yates correction for low expected frequencies) or 2-sample Student’s t-test as appropriate. The Pearson product moment correlation coefficient was used to measure linear correlations between variables. For correlation analysis, the log of NT-proBNP was used to normalize the distribution area. Correlations were compared by a z-test on normally transformed correlation coefficients. The relative abilities of NT-proBNP to predict the selection criteria for cardiac transplantation were assessed by receiver operating characteristic (ROC) analysis. AUC for each marker was compared by the method of Hanley and McNeil.21 Decision statistics were computed from 2 ⫻ 2 tables and reported as sensitivity, specificity, and positive and negative predictive values. Diagnostic accuracy was computed as the sum of the concordant cells divided by the sum of all cells in the 2 ⫻ 2 table. Agreement between clinical judgment and
Rothenburger et al.
1191
NT-proBNP was quantified by Cohen’s statistic. The positive likelihood ratio was taken as the slope of the ROC curve for the optimum cutpoint and was expressed as follows: (sensitivity/1 ⫺ specificity). Multivariate analysis was performed by the Cox binary logistic regression model. Statistical significance was assumed for a p value less than 0.05. RESULTS Baseline Characteristics A total of 550 patients were prospectively investigated. The demographics and baseline characteristics for the study sample according to the underlying heart disease are reported in Table 1. No differences between groups regarding weight, heart rate, ejection fraction, shortening fraction, mean arterial pressure, cardiac index, serum sodium, serum creatinine, and HFSS were observed. Patients with DCM were significantly taller (p ⬍ 0.01). Their ventricular diameters left ventricular enddiastolic diameter (LVEDD) and left ventricular endsystolic diameter (LVESD) were greater (p ⬍ 0.01), the cardiac output was higher (p ⫽ 0.041), and the maximum oxygen consumption was greater when compared with the ischemic patients (p ⫽ 0.004; Table 1). NT-proBNP and NYHA Class The baseline characteristics of patients grouped according to NYHA functional class are presented in Table 2. NT-proBNP levels were significantly higher with increasing NYHA class (p ⫽ 0.0024; Figure 1). Patients in NYHA Class 1 served as the reference group. In these 51 patients, the mean level of NT-proBNP was 50 ⫾ 13 ng/ml. The levels increased from NYHA Class 2 (150 ⫾ 56 ng/ml) to NYHA Class 3 (1800 ⫾ 452 ng/ml) and NYHA Class 4 (3800 ⫾ 499 ng/ml). NT-proBNP and HFSS The baseline characteristics by HFSS risk group are reported in Table 3. NT-proBNP levels were significantly higher in the HFSS high-risk group (p ⫽ 0.013; Figure 2). Low-risk patients had low NT-proBNP levels of (249 ⫾ 64 ng/ml), whereas the medium-risk patients (950 ⫾ 97 ng/ml) and high-risk patients (2689 ⫾ 552 ng/ml) showed significantly higher values. NT-proBNP and Selection of Transplant Candidates NT-proBNP levels of 254 patients assigned to heart transplantation were significantly higher than in 296 patients who were not assigned (2293 pg/ml vs 493 pg/ml; p ⬍ 0.001). In patients with DCM, the NT-proBNP levels were significantly higher in the group considered for heart transplantation (2789 ⫾ 467 ng/ml vs 280 ⫾ 123 ng/ml; p ⬍ 0.001; Figure 3). Patients with CAD also showed higher NT-proBNP levels in the group assigned for heart
1192
Rothenburger et al.
The Journal of Heart and Lung Transplantation
Table 2. Patient Characteristics by NYHA Functional Class* Characteristic No. of patients Age, y Sex (M/F) QRS ⬎0.12, % LV-EF, % LVEDD, mm LVESD, mm FS, % Mean arterial pressure, mm Hg Cardiac output, L/min Cardiac index, L/min/m2 Serum sodium, mg/dl Serum creatinine, mg/dl V˙ O2 max., ml/min/kg Heart Failure Survival Score
NYHA class 1 51 46 ⫾ 10 26/25 21,1 38 ⫾ 13 65 ⫾ 10 50 ⫾ 10 24 ⫾ 6 79 ⫾ 11 5.7 ⫾ 0.6 3.9 ⫾ 0.6 138 ⫾ 1 1.0 ⫾ 0.2 21.4 ⫾ 4.7 9.72 ⫾ 1
NYHA class 2 223 57 ⫾ 9 191/32 65.2 33 ⫾ 11 68 ⫾ 10 54 ⫾ 12 21 ⫾ 8 75 ⫾ 13 5.0 ⫾ 1 3 ⫾ 0.7 139 ⫾ 3 1.3 ⫾ 0.7 16.8 ⫾ 3.1 8.96 ⫾ 0.92
NYHA class 3 262 53 ⫾ 13 170/92 77.5 25 ⫾ 8 67 ⫾ 10 54 ⫾ 11 18 ⫾ 6 66 ⫾ 13 4.3 ⫾ 1 2.4 ⫾ 0.6 138 ⫾ 3 1.4 ⫾ 0.5 13.9 ⫾ 3.3 8.03 ⫾ 0.76
NYHA class 4 14 54 ⫾ 10 10/4 100 23 ⫾ 12 69 ⫾ 12 60 ⫾ 15 14 ⫾ 7 60 ⫾ 4 3.7 ⫾ 0.3 2.0 ⫾ 0.2 136 ⫾ 5 2.4 ⫾ 1.2 11.1 ⫾ 2.3 7.08 ⫾ 0.86
p Value ⬍0.001 ⬍0.001 ⬍0.01 NS NS ⬍0.001 ⬍0.01 ⬍0.001 ⬍0.001 ⬍0.02 ⬍0.001 ⬍0.001 ⬍0.001
*LV-EF, left ventricular ejection fraction; FS, shortening fraction; LVEDD, left ventricular enddiastolic diameter; LVESD, left ventricular endsystolic diameter.
transplantation (1800 ⫾ 430 ng/ml vs 156 ⫾ 68 ng/ml; p ⬍ 0.01). Results of the univariate and multivariate analysis of different parameters are shown in Table 4. Receiver Operating Characteristic Curves In decision making for or against heart transplantation, arbitrary cutoff points for NT-proBNP were used. For patients with DCM, an arbitrary cutoff point of 2000 pg/ml was defined. At this cutoff point, a positive predictive value of 0.87, a negative predictive value of 0.89, a sensitivity of 0.84, and a specificity of 0.88 were found, resulting in an odds ratio of 7.65 (p ⬍ 0.001). Compared with NYHA and HFSS, NT-proBNP the results
were significantly better (p ⬍ 0.001). For patients with CAD, an arbitrary cutoff point of 1000 pg/ml was defined. At this cutoff point, a positive predictive value of 0.96, a negative predictive value of 0.87, a sensitivity of 0.84, and a specificity of 0.96 were found, resulting in an odds ratio of 5.14 (p ⬍ 0.001). Compared with NYHA and HFSS, NT-proBNP the results were significantly better (p ⬍ 0.001). By use of an arbitrary cutoff point of 2000 mg/ml, regarding the decision for heart transplant candidacy showed, the use of the HFSS only revealed an AUC of 0.847 ⫾ 0.017 (range 0.814 – 0.880; Figure 4). Making decisions that used the NYHA classification only showed an AUC of 0.859 ⫾ 0.017 (range 0.825– 0.892), and the use of NT-proBNP 0.962 ⫾ 0.008 (range 0.945– 0.978; p ⬍ 0.001 for pairwise comparisons; Figure 4). The combined use of all 3 parameters improved the AUC to 0.984 ⫾ 0.012 (range 0.944 – 0.989) and therefore the significance of the decision for heart transplant candidacy. NT-proBNP as a Predictive Marker for Treatment Consequences
Figure 1. NT-proBNP in correlation to New York Heart Association (NYHA) functional classification; *p ⬍ 0.001 significances NYHA Class 3 and 4 vs other groups. BNP, B-type natriuretic peptide.
The mean NT-proBNP levels of all patients NYHA Class 2 or more (n ⫽ 276, 50.2%) who were admitted for the first time to our interdisciplinary heart failure office were 2500 ⫾ 560 pg/ml. As a consequence of optimized drug treatment among NYHA Class 3 and 4 patients, NT-proBNP levels decreased (3 ⫾ 1 month) in 187 patients (67.7%; NT-proBNP 876 ⫾ 279 pg/ml vs 2670 ⫾ 564 pg/ml; p ⫽ 0.023). NT-proBNP remained elevated or increased in 89 patients (32.3%; NT-proBNP 3217 ⫾ 829 pg/ml vs 2760 ⫾ 564 pg/ml; p ⫽ 0.03). Of these, 10 patients were treated immediately with a left ventricular assist device (LVAD).
The Journal of Heart and Lung Transplantation
Rothenburger et al.
1193
Table 3. Patient Characteristics by HFSS Risk Group* HFSS Characteristic No. of patients Age, years Sex (M/F) QRS ⬎ 0.12, % NYHA class LV-EF, % LVEDD, mm LVESD, mm FS, % Mean arterial pressure, mm Hg Cardiac output, L/min Cardiac index, L/min/m2 Serum sodium, mg/dl Serum creatinine, mg/dl V˙ O2 max., ml/min/kg Heart Failure Survival Score
High risk 58 58 ⫾ 10 42/16 91.6 3⫾0 22 ⫾ 4 72 ⫾ 9 59 ⫾ 9 15 ⫾ 4 60 ⫾ 7 3.7 ⫾ 1.2 1.9 ⫾ 0.6 137 ⫾ 3 1.7 ⫾ 0.8 11.9 ⫾ 3 6.95 ⫾ 0.27
Medium risk 166 54 ⫾ 9 128/38 85.7 3⫾1 22 ⫾ 4 70 ⫾ 7 58 ⫾ 8 17 ⫾ 5 62 ⫾ 7 4.4 ⫾ 1 2.5 ⫾ 0.7 138 ⫾ 3 1.4 ⫾ 0.7 14.5 ⫾ 3.7 7.76 ⫾ 0.28
Low risk 326 54 ⫾ 13 224/102 26.2 2⫾1 35 ⫾ 11 65 ⫾ 10 52 ⫾ 12 22 ⫾ 8 76 ⫾ 14 5.0 ⫾ 0.9 3.0 ⫾ 0.7 139 ⫾ 2 1.3 ⫾ 0.5 16.9 ⫾ 3.8 9.16 ⫾ 0.81
p Value NS ⬍0.02 ⬍0.01 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.001 ⬍0.01 ⬍0.001 ⬍0.001 NS NS ⬍0.001 ⬍0.001
*LV-EF, left ventricular ejection fraction; FS, shortening fraction; LVEDD, left ventricular enddiastolic diameter; LVESD, left ventricular endsystolic diameter.
Two patients underwent urgent heart transplantation. An implantable cardioverter system (ICD) was implanted in 39 patients. NT-proBNP as a Prognostic Marker in CHF Patients The NT-proBNP levels above 5000 pg/ml were in 58 patients (21% of NYHA Class 3 and 4 patients). The mortality rate among NYHA Class 3 and 4 patients was 10.2% (n ⫽ 28). The survivors in this group (30 of 58 patients) were treated with either LVAD support (n ⫽
Figure 2. NT-proBNP and correlation to risk groups of Heart Failure Survival Score; significance high risk group vs other groups *p ⬍ 0.01. BNP, B-type natriuretic peptide.
10) or urgent heart transplantation (n ⫽ 2). In 18 patients, an ICD system implantation were performed. During the follow-up period of 2 years, 28 of 276 NYHA Class 3 and 4 patients (10.2%) died while on the waiting list for heart transplantation (n ⫽ 9) or during evaluation for heart transplantation (n ⫽ 19). The reasons of death were heart failure (n ⫽ 17), severe arrhythmias (n ⫽ 8), and noncardiac causes (n ⫽ 3). The mean NT-proBNP
Figure 3. NT-proBNP levels in patients with chronic heart failure assigned to cardiac transplantation or not; *p ⬍ 0.001 significance of decision differences for each group. BNP, B-type natriuretic peptide; DCM, dilative cardiomyopathy; CAD, coronary artery disease.
1194
Rothenburger et al.
The Journal of Heart and Lung Transplantation
Table 4. Factors for Decision Making for Cardiac Transplantation* Variable Age (⬎60 years) Weight (⬎85 kg; ⬍40 kg) Height (⬎180 cm; ⬍120 cm) CAD HFSS (⬍ 8.09) NYHA class (3/4) NT-proBNP (⬎1.000 ng/ml) V˙ O2max (⬍12 ml/min/kg) Creatinine (⬍1.5 mg/dl) Sodium (⬍133 mmol/l) Heart rate (⬎100 bpm) QRS (⬎0.12ms) CO (⬍4.0 L/min) CI (⬍2.0 L/min/m2) EF (⬍30%) FS (⬍25%) LVEDD (⬎ 6 cm) LVESD (⬎4 cm)
p Value (univariate) 0.237 0.017 0.314 0.012 ⬍0.001 ⬍0.001 ⬍0.001 0.042 0.404 0.321 0.052 0.023 0.962 0.051 0.031 0.364 0.779 0.529
OR (CI 95%) 0.48 (0.28–7.22) 3.47 (1.94–8.23) 0.66 (0.12–8.49) 2.4 (1.01–7.43) 7.42 (4.5–9.87) 8.48 (3.9–11.9) 10.6 (3.7–14.5) 2.07 (1.26–5.11) 1.54 (0.4–7.62) 1.29 (0.07–8.23) 2.68 (1.2–7.53) 2.47 (1.26–5.11) 1.08 (0.4–9.32) 1.45 (0.62–5.8) 2.21 (1.03–7.67) 1.32 (0.79–3.45) 1.08 (0.7–6.31) 1.12 (0.4–7.21)
p Value (multivariate) 0.28 0.008 0.26 0.039 0.02 0.01 0.01 0.58 0.12 0.19 0.15 0.46 0.90 0.56 0.02 0.41 0.80 0.48
*CI, confidence interval; OR, odds ratio; CAD, coronary artery disease; CO, cardiac output; CI, cardiac index; EF, ejection fraction; FS, shortening fraction; LVEDD, left ventricular enddiastolic diameter; LVESD, left ventricular endsystolic diameter.
level in these patients was 5423 ⫾ 423 pg/ml. The mortality rate significantly correlated with NT-proBNP levels (r ⫽ 0.71; p ⬍ 0.02). Patients with NT-proBNP levels below 1000 pg/ml at admission had a yearly mortality rate of 0.7%; for levels of 1000 to 4999 pg/ml, the mortality rate was 1.6%; and for levels above 5000 pg/ml, the mortality rate was 28.4% (p ⫽ 0.014).
DISCUSSION The poor prognosis associated with advanced heart failure has been demonstrated by Gardner and coworkers.22 They described a 1-year mortality of approximately 15% among patients suffering from advanced heart failure. The recent COPERNICUS Study has provided comparable 1-year mortality rates of 18.5% in the placebo-treated group and 11.4% in the carvediloltreated group.23,24 The high mortality rates of end-stage heart failure make it desirable that parameters be found that help identify patients at high risk of death or adverse outcome. These patients can then be considered for cardiac transplantation or other surgical treatment, such as insertion of LVADs. The present study provides a large single-center comparison of prognostic assessment by use of HFSS in patients with heart failure due to different causes. Our study demonstrates that the measurement of NTproBNP concentrations may serve as an independent marker of mortality, and it provides important information for clinical decision making in heart transplant or LVAD candidates. The study further provides the predictive and prognostic impact of NT-proBNP levels on treatment options and mortality of CHF patients. NT-proBNP and Etiology
Figure 4. Receiver operating characteristic (ROC) curves for the sensitivity and specificity of NT-proBNP, Heart Failure Survival Score (HFSS), and New York Heart Association (NYHA) class and assignment to heart transplantation; (p ⬍ 0.001 for all pairwise comparisons of area under ROC curve). BNP, B-type natriuretic peptide.
The levels of NT-proBNP in patients with DCM were significantly higher than in patients with left ventricular dysfunction due to coronary artery disease. There was a strong correlation between NYHA class and HFSS to NT-proBNP. Potential reasons why increased left ven-
The Journal of Heart and Lung Transplantation
tricular diameters and ventricular wall tension results in significantly higher NT-proBNP levels in patients with DCM have been discussed.13,22 Consensus exists that increasing levels of NT-proBNP strongly correlate with both increased ventricular chamber diameters and increased ventricular wall tension. Recent reports indicate that cardiac neurohormone levels reflect left ventricular function. They add useful information beyond clinical features, including results of cardiac imaging, in staging and predicting outcome of patients with CHF due to dilated cardiomyopathy or after myocardial infarction.9,25,26 As a putative measure of the hemodynamic consequences of heart failure in myocardial ischemia or after myocardial infarction, BNP has been shown to be useful for risk assessment among patients with non-ST elevation acute coronary syndrome.1,26 Fewer data are available in advanced left ventricular dysfunction due to coronary artery disease. The impact of NT-proBNP in patients after multiple myocardial infarctions without options for revascularization, even without symptoms such as angina pectoris, have not been elucidated. However, NT-proBNP is markedly lower in left ventricular dysfunction due to coronary artery disease compared with DCM, despite similar ejection fraction or NYHA class. The results of our study demonstrate a close correlation to NYHA classification and HFSS, and confirm the prognostic impact of NT-proBNP in patients with advanced heart failure. NT-proBNP, and HFSS and NYHA Functional Class Heart failure can be difficult to diagnose accurately because the signs and symptoms of this disorder are neither sensitive nor specific. These limitations are especially relevant when symptoms are mild, or when patients are elderly or have comorbid disorders that mimic heart failure, such as pulmonary disease or obesity. Studies have shown that BNP may help to distinguish patients suffering from dyspnea due to heart failure or pulmonary reasons.2,5–7,17 By use of a BNP threshold of more than 100 pg/ml to diagnose heart failure, BNP did better than all other clinical variables and the clinical judgment of the emergency room physician. It contributed explanatory power to a multivariable model that incorporated diagnostic variables from the patient’s history, examination, and chest x-ray.2,5–7 Asymptomatic left ventricular dysfunction is at least as common as symptomatic heart failure. A simple screening test might help identify patients at risk of developing heart failure who would benefit from treatment that prevents progression to heart failure, including angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, and -blockers. Such screening could be targeted to patients at high risk for left
Rothenburger et al.
1195
ventricular dysfunction, such as those with diabetes, recent myocardial infarction, or end-stage renal failure.27–29 NT-proBNP and Decision Making in Selection of Heart Transplant Candidates Beyond its diagnostic and prognostic value, NT-proBNP can also help physicians in clinical decision making, particularly with regard to the question of transplant candidacy in patients with advanced heart failure. The premise of this approach is that decisions to initiate or titrate pharmacological treatments, or to use a more invasive strategy such as cardiac transplantation, might be based not only on symptoms and physical examination findings, but also on plasma levels of NT-proBNP. Thus, NT-proBNP could ultimately prove useful in helping physicians to select the appropriate drugs and drug doses.30 Further data are needed to demonstrate the potential value of NT-proBNP to assess the need for more invasive strategies such as implantable defibrillators, ventricular assist devices, or cardiac transplantation. The present study shows a good correlation of the levels of the neurohormone NT-proBNP and the consideration for cardiac transplantation. On the basis of clinical findings, high levels of NT-proBNP were strongly associated with a high risk of fatal outcome. On the basis of these findings, we suggest an arbitrary cutoff point of 2000 pg/ml for patients with DCM and of 1000 pg/ml for patients with severe left ventricular dysfunction due to coronary artery disease. More than 90% of all patients above these cutoff points of NTproBNP were clinically considered for heart transplant candidacy when examined by blinded cardiologists. Because these values showed a high sensitivity and specificity in decision making in heart transplant candidates, noninvasive NT-proBNP may help unravel the complex problem of whom to refer for cardiac transplantation. NT-proBNP as a Predictive Marker for Treatment Consequences Decision making for different treatment options might sometimes be difficult in CHF patients. In the present study, NT-proBNP provides a useful parameter in controlling treatment effects. Patients who initiate drug treatment could be followed up regarding treatment success, which results in a decrease of NT-proBNP levels. Although an improvement in NYHA classification and HFSS provides similar informations, the first one is subjective, and the second one requires many examinations before a decision is possible. Patients without any changes or increases in NT-proBNP due to drug treatment should be undergo either LVAD implantation or urgent heart transplantation.
1196
Rothenburger et al.
NT-proBNP as a Prognostic Marker In this study, NT-proBNP has been detected as an prognostic marker in CHF patients. Levels above 5000 pg/ml were associated with a markedly increased mortality rate. The intervention rate in these patients was significantly higher; most of them underwent LVAD implantation, urgent heart transplantation, or implantation of an ICD. Tsutamoto and coworkers31 reported that natriuretic peptides offer a new option in evaluating prognosis in patients with clinical heart failure. In contrast, Berger et al32 showed that big endothelin seems to be a better marker to estimate the prognosis of CHF than neuroendocrine hormones within the first year. Selvais et al33 reported that endothelin was superior to BNP in predicting 3-year mortality in a population of 109 patients with symptomatic heart failure. Nevertheless, our data provide NT-proBNP levels in CHF patients that have not been shown in terms of the prognostic value of this marker in previous studies. In our study, NT-proBNP was found to be highly associated with prognosis in CHF patients. Conclusions NT-proBNP was found to be a helpful marker in selection of heart transplant candidates. Furthermore, NTproBNP provides an effective parameter of treatment control in CHF patients. It is useful in decision making for further treatment options (medically, LVAD implantation or urgent heart transplantation) in CHF patients. The prognostic value of NT-proBNP may be helpful in finding patients for require immediate hospitalization for heart failure. REFERENCES 1. DeLemos JA, Morrow DA, Bentley JH, et al. The prognostic value of B-type natriuretic peptide in patients with acute coronary syndromes. N Engl J Med 2001;345:1014 – 21. 2. 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. 3. McCullough PA, Nowak RM, McCord J, et al. B-type natriuretic peptide and clinical judgement in emergency diagnosis of heart failure. Circulation 2002;106:416 –22. 4. Mair J, Hammerer-Lercher A, Puschendorf B. The impact of cardiac natriuretic peptide determination on the diagnosis and management of heart failure. Clin Chem Lab Med 2001;39:571– 88. 5. Vesely DL. Natriuretic peptides and acute renal failure. Am J Physiol Renal Physiol 2003;285:F167–77. 6. Maisel AS, McCord J, Nowak RM, et al. Bedside B-type natriuretic peptide in the emergency diagnosis of heart failure with reduced or preserved ejection fraction. J Am Coll Cardiol 2003;41:2010 –7.
The Journal of Heart and Lung Transplantation
7. Morrison LK, Harrison A, Krishnaswamy P, et al. Utility of a rapid B-natriuretic peptide assay in differentiating congestive heart failure from lung disease in patients presenting with dyspnoea. J Am Coll Cardiol 2002;39:202–9. 8. Aaronson KD, Schartz JS, Chen TM, et al. Development and prospective validation of a clinical index to predict survival in ambulatory patients referred for cardiac transplant evaluation. Circulation 1997;95:2660 –7. 9. Richards AM, Nicholls MG, Yandle TG, et al. Neuroendocrine prediction of left ventricular function and heart failure after acute myocardial infarction. Heart 1999;81: 114 –20. 10. Schrier RW, Abraham WT. Hormones and hemodynamics in heart failure. N Engl J Med 1999;341:577– 85. 11. Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med 1998;339:321– 8. 12. De Lemos JA, McGuire DK, Drazner MH. B-type natriuretic peptide in cardiovascular disease. Lancet 2003;362: 316 –22. 13. Ruskoaho H. Cardiac hormones as diagnostic tools in heart failure. Endocr Rev 2003;24:341–56. 14. Jourdain P, Funck F, Bellorini M, et al. Bedside B-type natriuretic peptide and functional capacity in chronic heart failure. Eur J Heart Failure 2003;5:155– 60. 15. Koch A, Singer H. Normal values of B type natriuretic peptide in infants, children, and adolescents. Heart 2003; 89:875– 8. 16. Raymond I, Groenning BA, Hildebrandt PR, et al. The influence of age, sex, and other variables on the plasma level of N-terminal pro brain natriuretic peptide in a large sample ot the general population. Heart 2003;89:745–51. 17. Goetze JP, Videbaek R, Boesgaard S, et al. Pro-brain natriuretic peptide as marker of cardiovascular or pulmonary causes of dyspnea in patients with terminal parenchymal lung disease. J Heart Lung Transplant 2004;23: 80 –7. 18. Richards AM, Nicholls MG, Espiner EA, et al. B-type natriuretic peptides and ejection fraction for prognosis after myocardial infarction. Circulation 2003;107:2786 – 92. 19. Wieczorek SJ, Wu AHB, Christenson R, et al. A rapid B-type natriuretic peptide assay accurately diagnoses left ventricular dysfunction and heart failure: a multicenter evaluation. Am Heart J 2002;144:834 –9. 20. Cimato TR, Jessup M. Recipient selection in cardiac transplantation: contraindications and risk factors for mortality. J Heart Lung Transplant 2002;21:1161–73. 21. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143:29 –36. 22. Gardner RS, Özalp F, Murday AJ, et al. N-terminal probrain natriuretic peptide. A new gold standard in predicting mortality in patients with advanced heart failure. Eur Heart J 2003;24:1735– 43. 23. Packer M, Coats AJ, Fowler MB, et al. Effect of carvedilol on survival in severe chronic heart failure. N Engl J Med 2001;344:1651– 8. 24. Packer M, Fowler MB, Roecker EB, et al. Effect of carvedilol on the morbidity of patients with severe chronic heart failure: results of the Carvedilol Prospective
The Journal of Heart and Lung Transplantation
25.
26.
27.
28.
29.
Randomized Cumulative Survival (COPERNICUS) study. Circulation 2002;106:2194 –9. Fisher C, Berry C, Blue L, et al. N-terminal pro B type natriuretic peptide, but not the new putative cardiac hormone relaxin, predicts prognosis in patients with chronic heart failure. Heart 2003;89:879 – 81. Goetze JP, Christoffersen C, Perko M, et al. Increased cardiac BNP expression associated with myocardial ischemia. FASEB J 2003;17:U24 –33. Struthers AD, Morris AD. Screening for and treating left-ventricular abnormalities in diabetes mellitus: a new way of reducing cardiac death. Lancet 2002;359:1430 –2. Motwani JG, McAlpine H, Kennedy N, et al. Plasma brain natriuretic peptide as an indicator for angiotensin-converting enzyme inhibition after myocardial infarction. Lancet 1993;341:1109 –13. Mallamaci F, Zoccali C, Tripepi G, et al. Diagnostic potential of cardiac natriuretic peptides in dialysis patients. Kidney Int 2001;59:1559 – 66.
Rothenburger et al.
1197
30. Richards AM, Doughty R, Nicholls MG, et al. Plasma N-terminal pro-brain natriuretic peptide and adrenomedullin: prognostic utility and prediction of benefit from carvedilol in chronic ischemic left ventricular dysfunction. J Am Coll Cardiol 2001;37:1781–7. 31. Tsutamoto T, Wada A, Maeda K, et al. Attenuation of compensation of endogenous cardiac natriuretic peptide system in chronic heart failure: prognostic role of plasma brain natriuretic peptide concentration in patients with chronic symptomatic left ventricular dysfunction. Circulation 1997;96:509 –16. 32. Berger R, Strecker K, Huelsmann M, et al. Prognostic power of neurohormonal parameters in chronic heart failure depends on clinical stage and observation period. J Heart Lung Transplant 2003;22:1037– 45. 33. Selvais PL, Robert A, Ahn S, et al. Direct comparison between endothelin-1, N-terminal proatrial natriuretic factor, and brain natriuretic peptide as prognostic markers of survival in congestive heart failure.