Received 24 January 2007; received in revised form 27 February 2007; accepted 5 July 2007

Transcription

1 European Journal of Heart Failure 9 (2007) Prognostic importance of plasma NT-pro BNP in chronic heart failure in patients treated with a β-blocker: Results from the Carvedilol Or Metoprolol European Trial (COMET) trial Lars G. Olsson a, Karl Swedberg a,, John G.F. Cleland b, Phillip A. Spark c, Michel Komajda d, Marco Metra e, Christian Torp-Pedersen f, Willem J. Remme g, Armin Scherhag h, Philip Poole-Wilson i on behalf of the COMET investigators a Department of Emergency and Cardiovascular Medicine Sahlgrenska Academy, Göteborg University, Göteborg, Sweden b University of Hull, Kingston upon Hull, United Kingdom c Nottingham Clinical Research Group, Nottingham, United Kingdom d Department of Cardiology, La Pitié-Salpétrière Hospital, Paris France e Section of Cardiovascular Diseases, Department of Experimental and Applied Medicine, University of Brescia, Italy f Department of Cardiology, Bispebjerg University Hospital, Copenhagen, Denmark g Sticares Cardiovascular Research Institute, Rhoon, The Netherlands h Medical Clinic, University Hospital Mannheim, University of Heidelberg, Germany i National Heart and Lung Institute, Imperial College, London, United Kingdom Received 24 January 2007; received in revised form 27 February 2007; accepted 5 July 2007 Abstract Background: Plasma levels of N-terminal pro-brain natriuretic peptide (NT-pro BNP) are increased in patients with chronic heart failure (CHF). Beta-blockers (BB) may influence these levels but it is unclear whether changes in NT-pro BNP reflect concomitant changes in prognosis. Objectives: To assess the prognostic importance of NT-pro BNP at baseline and during follow-up, in patients in whom beta-blocker therapy is initiated. Methods: In COMET, 3029 patients with CHF in NYHA class II IV and EFb35% were randomised to carvedilol or metoprolol tartrate and were followed for an average of 58 months. Blood samples were collected for the measurement of NT-pro BNP at baseline (n=1559) and during follow-up (n = 309). Results: Baseline plasma concentrations of NT-pro BNP above the median (1242 pg/ml) were associated with higher all-cause mortality (RR 2.77; 95% CI , pb0.001). Patients who achieved NT-pro BNP levels b400 pg/ml during follow-up had a lower subsequent mortality (RR 0.32; 95% CI , p=0.004). Conclusions: The plasma concentration of NT-pro BNP is a powerful predictor of mortality in patients with CHF. Patients who achieve an NT-pro BNP of b400 pg/ml subsequent to treatment with a beta-blocker have a favourable prognosis European Society of Cardiology. Published by Elsevier B.V. All rights reserved. Keywords: Chronic heart failure; Natriuretic peptides; Prognosis; Treatment Abbreviations: ANP, Atrial Natriuretic Peptide; BATTLESCARRED, NT-pro BNP-Assisted Treatment Lessen Serial Cardiac Readmission And Death; COMET, Carvedilol Or Metoprolol European Trial; COPERNICUS, Carvedilol Prospective Randomized Cumulative Survival Study Group; NT-pro BNP, N- amino terminal pro brain natriuretic peptide; PRAISE-2, Second Prospective Amlodipine Survival Evaluation; Val-HeFT, Valsartan in Heart Failure Trial. Corresponding author. Department of Medicine, Sahlgrenska University Hospital/Östra, SE Göteborg, Sweden. Tel.: ; fax: address: karl.swedberg@gu.se (K. Swedberg) /$ - see front matter 2007 European Society of Cardiology. Published by Elsevier B.V. All rights reserved. doi: /j.ejheart

2 796 L.G. Olsson et al. / European Journal of Heart Failure 9 (2007) Introduction Chronic heart failure (CHF) is a syndrome involving not just the cardiovascular system but almost all body systems. Ventricular dysfunction leads to activation of the neuroendocrine system in an attempt to correct haemodynamic disturbances but this leads to vasoconstriction, increases in heart rate and salt and water retention. The onset of overt heart failure indicates that compensatory mechanisms have been over-whelmed. Continued neuroendocrine activation may not only be a useful marker of the severity of heart failure but may also mediate further cardiovascular damage and is a marker of an adverse outcome [1,2]. Inhibition of neuroendocrine activation by blocking the renin angiotensin aldosterone system and the sympathetic system has been effective in reducing morbidity and mortality [3]. The value of prognostic markers identified prior to recent therapeutic developments needs to be re-assessed. Pre-pro brain natriuretic peptide (BNP) is synthesized in the myocardium and split into a physiologically active component, BNP with 32 amino acids and a relatively short half-life and the much more long-lived inactive metabolite NT-pro BNP with 76 amino acids [4,5]. The latter peptide is stable invitro at room temperature which allows it to be used as a robust routine clinical tool. Biologically active BNP and the inactive fragment NT-pro BNP both reflect haemodynamic [6] and myocardial function [7] and their levels are strong markers of cardiac remodelling [8], both can be used to exclude a diagnosis of CHF, as a prognostic marker [9,10] and to predict the response to therapy [11]. Some effective treatments for heart failure, such as angiotensin converting enzyme (ACE) inhibitors and cardiac resynchronization therapy can reduce plasma concentrations of natriuretic peptides, suggesting that natriuretic peptides might be used to guide treatment intensity [12]. Beta-blockers improve prognosis but their effects on natriuretic peptides are controversial, with studies variously showing an increase, decrease or no change [13 16], leaving doubts about the prognostic value of natriuretic peptides in patients treated with a beta-blocker [17]. In COMET (Carvedilol Or Metoprolol European Trial) patients with CHF were randomised to treatment with one of two beta-blockers and treated for an average of 58 months [18]. In a large proportion of COMET patients, blood samples for assessment of NT-pro BNP were collected both at baseline and during follow-up, in order to assess (1) the prognostic importance of NT-pro BNP measured at baseline in patients subsequently receiving beta-blocker therapy and (2) the prognostic importance of changes in NT-pro BNP levels. 2. Methods The COMET study was a randomised, double blind comparison of carvedilol with metoprolol tartrate. A detailed description of the study design and inclusion/exclusion criteria has been published [19]. In summary,eligible patients had symptomatic chronic heart failure (New York Heart Association (NYHA) class II IV) and at least one cardiovascular admission during the previous 2 years. Left-ventricular ejection fraction had to be b 0.35 measured within the previous 3 months. Major exclusion criteria were: requirement for intravenous inotropic therapy, current treatment with calcium channel blockers, amiodarone (N 200 mg/day), class-i antiarrhythmic drugs, unstable angina, myocardial infarction, coronary revascularisation or stroke within the previous 2 months. At randomisation, 3029 patients were allocated to receive either mg carvedilol twice daily or 5 mg metoprolol tartrate twice daily; with dose doubling bi-weekly until maximally tolerated or the target dose of carvedilol 25 mg twice daily or metoprolol 50 mg twice daily was achieved. In a sub-set of patients, blood samples for the measurement of NT-pro BNP concentrations were collected at baseline and during follow-up. The first follow-up sample was planned at one year and another sample one year later. However, due to varying levels of participation in this sub-study, 1559 samples were collected at baseline and 528 samples at varying timepoints during follow-up. In order to minimise variation in the duration of follow-up, we restricted our analyses to patients with a follow-up sample collected months after baseline (n = 309). This analysis therefore focuses on the 1559 patients with baseline samples and the 309 patients with follow-up samples within the afore-mentioned time frame. NT-pro BNP plasma concentrations were determined in a core laboratory using the commercially available Elecsys 2010 automated immunoassay (Roche Pharmaceuticals) [20]. We evaluated follow-up changes by both median levels and percent change, due to the large and skewed distribution of NT-pro BNP values. In addition, we wanted a clinically meaningful cut-off level for follow-up assessment. We chose a concentration of 400 pg/ml as the cut-off level, as this is above the reference level for NT-pro BNP even for elderly women of 359 pg/ml [21], is lower than levels usually found in patients with CHF [22] and provides a simple, clinically meaningful guide. It was considered important to use one level instead of several age and sex dependent values. 3. Statistical analysis Differences in baseline characteristics between patients with or without baseline samples available were made using chi-squared tests for categorical data and t-tests for continuous parameters. Time to event analyses were assessed using Cox proportional hazard models, including randomised study medication in all models. Time to event analyses began at randomisation with the exception of those analyses of events post follow-up, which began at the time of the follow-up sample. In order to assess the impact of baseline characteristics on our results, we developed a multi-variable Cox proportional hazards model for all-cause mortality using bootstrap methods. As a form of internal validation, 200 samples from the 3029 patients (sampling with replacement) were taken and a backwards stepwise procedure run on each

3 L.G. Olsson et al. / European Journal of Heart Failure 9 (2007) Table 1 Baseline characteristics NT-proBNP sub-study population vs. all other patients in the main trial Baseline sample No baseline sample Total p value N=1559 N=1470 N=3029 Age (years) mean/sd 61.6/ / / Sex (% male) Systolic BP (mm Hg) 126.0/ / 126.1/ mean/sd 19.7 Diastolic BP (mm Hg) 77.1/ / / mean/sd Heart rate (bpm) mean/sd 81.2/ / / NYHA class Carmi (1990) % II % III % IV Duration CHF (months) 42.6/ / / mean/median Aetiology CHF a % Ischaemic heart disease % Hypertension % Dilated b cardiomyopathy LVEF mean/sd 25.7/ / / Previous MI % CAD (confirmed by angiography) % Current angina % Previous angioplasty % Previous CABG % Hypertension % Diabetes % Stroke % ECG findings % Sinus rhythm % Atrial fibrillation/ flutter % Paced rhythm Concomitant medication at randomisation Diuretics b % ACE inhibitors b % Angiotensin receptor antagonists % Digitalis % Antiarrhythmics % Nitrates % Aldosterone antagonists % Beta-blockers c % Anticoagulants % Aspirin % Lipid lowering agents (statins) % Laboratory measurements Haemoglobin level (g/dl) 14.2/ / / Serum creatinine (μmol/l) 107.2/ / 107.2/ Blood glucose (mmol/l) 6.6/ / / a More than one answer possible. b Inclusion criteria. c Stopped prior to study start. sample. For each sample, the variables selected were recorded. Variables appearing in 70% or more of all models are included in our final multivariate model. This included variables for study medication, age, sex, systolic blood pressure, NYHA class, duration of heart failure, left ventricular ejection fraction, electrocardiographic results, diuretic dose, use of digitalis, use of lipid lowering agents, and levels of haemoglobin, sodium and creatinine. The prognostic significance of NT-proBNP over time was also analysed using a time-dependent Cox regression analysis using the same set of variables with the follow-up NT-proBNP value included as a time-dependent covariate. All tests performed were two-sided and the significance level was No attempt has been made to adjust the significance level of the data presented for multiple testing. 4. Results A blood sample was collected at baseline in 1559 patients. Demographics for these patients and a comparison with the main study population are presented in Table 1. Patients included in the sub-study were more often in NYHA IV, more often had heart failure of ischaemic or hypertensive origin and were more often treated with nitrates and less often with anti-arrhythmic agents. The median plasma concentration of NT-pro BNP at baseline was 1242 pg/ml with a wide distribution among study participants (Fig. 1). Median baseline NT-pro BNP levels were 1185 pg/ml for patients assigned to metoprolol and 1298 pg/ml in those assigned to carvedilol. Mortality rates for patients with NT-pro BNP levels above vs. below median concentration were 52% and 24% respectively (relative risk (RR) 2.77 (95% CI: ; pb0.0001). In Fig. 2, mortality risks are presented by decile of NT-pro BNP plasma concentration, a progressive mortality risk with increasing BNP levels can be seen. In the highest decile compared to the lowest decile, N5146 vs. 236 pg/ml, RR was (95% CI ; p b0.0001). In the multivariable analysis, NT-pro BNP remained a strong, Fig. 1. Distribution of NT-pro BNP concentrations at baseline.

4 798 L.G. Olsson et al. / European Journal of Heart Failure 9 (2007) Fig. 2. Relative risk for mortality, and number of events by decile NT-pro BNP plasma concentration at baseline. independent predictor of all-cause mortality regardless of whether analysed by median values or deciles. In the 309 patients with both a baseline and medium-term follow-up sample available, the follow-up sample was obtained after 845 ± 181 days. The median (interquartile range, IQR) decrease in NT-pro BNP was 291 ( 73 to 976) pg/ml, the median change in percent was 37% and the median concentration achieved was 603 (201 to 1546) pg/ml. The median reduction in NT-pro BNP was 319 ( 66 to 1036) pg/ml in patients treated with metoprolol and 235 ( 96 to 964) pg/ml in patients treated with carvedilol. The achieved median concentration, but not the change in median level, predicted subsequent mortality: RR for levels median vs. b median 5.01 (95% CI ; p b ). The achieved median concentration was also of prognostic importance when treated as a time-dependent covariate in a multivariable analysis. Percentage reduction in NT-pro BNP less than the median change was associated with an increased risk of subsequent cardiovascular morbidity and mortality but not all-cause mortality (p = 0.067; Table 2). Patients who had NT-proBNP below the arbitrarily chosen cut off level 400 pg/ml at baseline had a markedly lower mortality than those who did not (RR 0.35; 95% CI ; pb ) (Fig. 3). Fewer patients (n=309) had both baseline and follow-up samples available, which diminished statistical power but the observed trend in mortality for patients with values b400 pg/ml (RR 0.39; 95% CI ; p = 0.07) was consistent with data from the larger population. However, a follow-up NT-pro BNP concentration b400 pg/ml was associated with lower subsequent all-cause mortality (RR 0.32; 95% CI ; p=0.004). Patients with an NTpro BNP below 400 pg/ml at baseline (n=57) and rising above 400 pg/ml during follow-up (n=17) had a slightly worse prognosis. Patients with an NT-pro BNP level 400 pg/ml at baseline (n=252), reaching an NT-pro BNP level below 400 pg/ml during follow-up (n=81) had a better prognosis than those whose levels remained high (RR 0.35: CI ; p=0.017) (Fig. 4). Table 2 Events during follow-up according to median percent change in NT-proBNP concentrations Endpoint Events/ Nbmedian reduction (%) All-cause mortality 28/154 (18.2%) Death plus all-cause 95/154 hospitalisation (61.7%) Death plus 78/154 cardiovascular (50.6%) hospitalisation Events/ N median reduction (%) 17/155 (11%) 65/155 (41.9%) 43/155 (27.7%) Relative risk; 95% CI 1.76; 0.96 to ; 1.35 to ; 1.50 to 3.17 p-value b0.001 b0.001 Fig. 3. Mortality according to baseline NT-pro BNP concentration b or z400 pg/ml in the 1559 patients with a blood sample at baseline.

5 L.G. Olsson et al. / European Journal of Heart Failure 9 (2007) Fig. 4. Subsequent mortality in those patients with NT-pro BNP concentrations z400 pg/ml at baseline (N=252) and with a concentrationbvs. z 400 pg/ml at follow-up. When patients were divided according to median NT-pro BNP levels, there was a trend (p=0.052) for lower mortality on carvedilol compared to metoprolol in patients with levels above the median, but no differences below median. Such sub-group analysis should be interpreted with caution since only half of patients in the main study had a measurement of NT-proBNP which will have reduced statistical power. 5. Discussion Our results confirm previous reports regarding the increased in mortality amongst patients with CHF and elevated plasma levels of NT-pro BNP and extend this risk association to patients on chronic beta-blocker therapy. Furthermore and importantly, continued high levels of plasma NT-pro BNP at follow-up were associated with increased subsequent mortality. Patients with reductions in NT-pro BNP at follow-up had a relatively good prognosis. Several studies have found baseline NT-pro BNP to be related to prognosis in different heart failure populations [23 28] and to perform at least as well as BNP in this respect [22]. Currently, the recommended use for BNP and NT-pro BNP is as a diagnostic tool, with low levels making a diagnosis of heart failure highly unlikely [9]. BNP levels vary widely within patient groups with similar systolic dysfunction. Important confounders are level of diastolic dysfunction, valve abnormalities, creatinine clearance [29], age and sex [30] and presence of atrial arrhythmias [31]. An alternative use is serial measurement of NT-pro BNP in the individual patient to monitor clinical progress and treatment effects. Several studies have been published on the prognostic impact of serial BNP measurements, both in acute heart failure and in clinically stable patients, with somewhat conflicting results. Both Maeda et al. and O'Brien et al. found that pre-discharge rather than admission NT-pro BNP levels predicted short-term survival for patients admitted to hospital for acute heart failure [32,33]. In a small study using both admission and pre-discharge levels of BNP, Cheng et al. reported that both absolute BNP levels at baseline and followup, as well as change in levels and percent change during admission predicted death or 30-day hospital readmission [34]. In a study by Bettencourt et al., 156 patients admitted to hospital because of decompensated heart failure had admission and pre-discharge NT-pro BNP levels measured and were followed for at least six months. Both a non-significant decrease (b ±30%)andaN30% increase in NT-pro BNP were strongly and independently correlated with adverse outcome when compared with a reduction of 30% [35]. In a subgroup analysis of the PRAISE-2 study, 181 patients had baseline and follow-up values of BNP and other neurohormones taken [36]. Neither baseline nor 26-week-BNP predicted survival, but change in BNP at 26 weeks did. Bettencourt et al. also evaluated 84 patients referred to a specialized heart failure clinic with baseline and follow-up measurements of BNP and followed for 1190 days [37]. Increments in BNP and NYHA class independently predicted all-cause mortality. The Val-HeFT investigators studied the prognostic importance of serial BNP measurements and found no difference in outcome depending on change in BNP and norepinephrine. However, when changes were expressed as percent change in quartiles, they observed an incremental and strong risk of cardiovascular morbidity and mortality [38].In a second analysis from Val-HeFT, patients with BNP measurements were divided into groups according to whether they were considered to have a low or a high value in relation to the median and whether the value increased, decreased or was steady during follow-up [39]. More patients in the group remaining above the median BNP level during follow-up reached the primary endpoint of cardiovascular morbidity and mortality compared with those remaining in the low group (27 vs. 8%). Patients with low values at study start which increased above the median BNP level during follow-up had a prognosis similar to those patients with persistently high levels (26%). Both elevated BNP levels throughout the study and increasing levels during the study were independently associated with morbidity and mortality for up to 12 months. Patients whose values fell tended to do better. Our study extends these findings by showing the predictive value of serial measurement of NTpro BNP levels in patients with chronic heart failure treated with beta-blockers. We have also shown that reaching low absolute NT-pro BNP levels has greater prognostic value than relative changes from baseline. The differences compared with other studies may be due to differences in the patient population (i.e. inclusion of stable vs. decompensated heart failure), use of BNP or NT-proBNP, the intervention, study duration and importantly study power. The improved relationship between NT-pro BNP and prognosis over time in patients treated with a beta-blocker has been noted by others [13]. One can hypothesize that this effect reflects the extent to which the patient has responded to treatment. It is logical that prognosis is determined more accurately by a recent rather than remote assessment of LV

6 800 L.G. Olsson et al. / European Journal of Heart Failure 9 (2007) function. Like baseline NT-pro BNP concentrations, changes in NT-pro BNP values were very heterogeneous during the study, this reflects normal biomarker behaviour but is probably also related to the differences in followup time. Both when expressed as concentration differences and percentage changes, the median changes at follow-up were small when expressed as either mean value or percent. The value of NT-pro BNP-levels as a guide to treatment intensity has been evaluated previously, but only in a relatively small group of patients [12]. The ongoing BATTLESCARRED trial should provide more information in this context [40] Effect of beta-blockade on NT-pro BNP concentrations Beta-blocker therapy prolongs cardiac filling times and is associated with negative inotropic effects, which can affect left ventricular wall-stress and the secretion of BNP. In 100 patients treated with atenolol or placebo for 24 months, NT-pro BNP but not BNP levels were reduced with non-significant differences between atenolol and placebo [13]. Fung et al. evaluated 49 patients on background ACEinhibitor treatment who received 52 weeks treatment with either carvedilol or metoprolol tartrate and found markedly reduced NT-pro BNP and ANP levels with a non-significant increase in N-BNP after four weeks [14]. In the COPERNI- CUS neurohormonal sub-study, NT-pro BNP levels in patients treated with carvedilol were slightly increased one month after initiation, then decreased during the following six months, with hormone levels unaltered during follow-up in the placebo group [15]. In the CHRISTMAS study (n = 305), carvedilol increased BNP, NT-proBNP and NT-ANP over 6 months in patients with ischaemic left ventricular dysfunction [41]. In a recent study, 16 patients with mild, stable heart failure were randomised to either metoprolol succinate or placebo and followed for six weeks [16]. ANP, BNP and their N-terminal versions all increased during the study period. The main difference between these studies and ours is the long follow-up time, with on average more than two years between the first and second samples, thus ruling out the acute effects of betablockade on natriuretic peptide levels. There is a possibility of other reasons than the introduction of beta-blockers for the reduction in NT-pro BNP levels, as there was no control group in our study. However, the main purpose of this analysis was not to evaluate treatment effects. We cannot exclude that the rather small number of patients with follow-up samples and the variable time to follow-up blood sampling, could introduce bias although we tried to reduce this by limiting the follow-up re-sampling time to 1 3 years. 6. Conclusions The plasma concentration of NT-pro BNP is an important prognostic marker in patients with chronic heart failure in patients receiving standard treatment for heart failure including beta-blockers. Repeat measurements improve prognostic accuracy, suggesting that this peptide is a useful way to monitor patients' progress and providing circumstantial evidence that it might be used as a therapeutic target. Patients with persistent high NT-pro BNP levels after initiation of therapy should, based on our findings, be considered for further intervention using pharmacological agents or devices that have been proven to improve prognosis. References [1] Cohn JN, Levine TB, Olivari MT, et al. Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 1984;311: [2] Swedberg K, Eneroth P, Kjekshus J, Wilhelmsen L. Hormones regulating cardiovascular function in patients with severe congestive heart failure and their relation to mortality. CONSENSUS Trial Study Group. Circulation 1990;82: [3] Swedberg K. Importance of neuroendocrine activation in chronic heart failure. Impact on treatment strategies. Eur J Heart Fail 2000;2: [4] Hunt PJ, Yandle TG, Nicholls MG, Richards AM, Espiner EA. The amino-terminal portion of pro-brain natriuretic peptide (Pro-BNP) circulates in human plasma. Biochem Biophys Res Commun 1995;214: [5] Hunt PJ, Espiner EA, Nicholls MG, Richards AM, Yandle TG. The role of the circulation in processing pro-brain natriuretic peptide (pro BNP) to amino-terminal BNP and BNP-32. Peptides 1997;18: [6] Vesely DL, Douglass MA, Dietz JR, et al. Three peptides from the atrial natriuretic factor prohormone amino terminus lower blood pressure and produce diuresis, natriuresis, and/or kaliuresis in humans. Circulation 1994;90: [7] Talwar S, Siebenhofer A, Williams B, Ng L. Influence of hypertension, left ventricular hypertrophy, and left ventricular systolic dysfunction on plasma N terminal pro BNP. Heart 2000;83: [8] Crilley JG, Farrer M. Left ventricular remodelling and brain natriuretic peptide after first myocardial infarction. Heart 2001;86: [9] Swedberg K, Cleland J, Dargie H, et al. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005): The Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology. Eur Heart J 2005;26: [10] Doust JA, Pietrzak E, Dobson A, Glasziou P. How well does B-type natriuretic peptide predict death and cardiac events in patients with heart failure: systematic review. BMJ 2005;330:625. [11] Richards AM, Doughty R, Nicholls MG, et al. Neurohumoral prediction of benefit from carvedilol in ischemic left ventricular dysfunction. Australia New Zealand Heart Failure Group. Circulation 1999;99: [12] Troughton RW, Frampton CM, Yandle TG, Espiner EA, Nicholls MG, Richards AM. Treatment of heart failure guided by plasma aminoterminal brain natriuretic peptide (N-BNP) concentrations. Lancet 2000;355: [13] Stanek B, Frey B, Hülsmann M, et al. Prognostic evaluation of neurohumoral plasma levels before and during beta-blocker therapy in advanced left ventricular dysfunction. J Am Coll Cardiol 2001;38: [14] Fung JW, Yu CM, Yip G, et al. Effect of beta blockade (carvedilol or metoprolol) on activation of the renin angiotensin aldosterone system and natriuretic peptides in chronic heart failure. Am J Cardiol 2003;92: [15] Hartmann F, Packer M, Coats AJ, et al. NT-pro BNP in severe chronic heart failure: rationale, design and preliminary results of the COPERNI- CUS NT-pro BNP substudy. Eur J Heart Fail 2004;6: [16] Davis ME, Richards AM, Nicholls MG, Yandle TG, Frampton CM, Troughton RW. Introduction of metoprolol increases plasma B-type

7 L.G. Olsson et al. / European Journal of Heart Failure 9 (2007) cardiac natriuretic peptides in mild, stable heart failure. Circulation 2006;113: [17] The RESOLVD investigators. Effects of metoprolol CR in patients with ischemic and dilated cardiomyopathy : the randomized evaluation of strategies for left ventricular dysfunction pilot study. Circulation 2000;101: [18] Poole-Wilson PA, Swedberg K, Cleland JG, et al. Comparison of carvedilol and metoprolol on clinical outcomes in patients with chronic heart failure in the Carvedilol Or Metoprolol European Trial (COMET): randomised controlled trial. Lancet 2003;362:7 13. [19] Poole-Wilson PA, Cleland JG, Di Lenarda A, et al. Rationale and design of the Carvedilol Or Metoprolol European Trial in patients with chronic heart failure: COMET. Eur J Heart Fail 2002;4: [20] Karl J, Borgya A, Gallusser A, et al. Development of a novel, N- terminal-pro BNP (NT-pro BNP) assay with a low detection limit. Scand J Clin Lab Invest Suppl 1999;230: [21] Huber KR, Mostafaie N, Bauer K, et al. Concentrations of N-terminal pro-brain natriuretic peptide and troponin T in plasma of 75-year-old apparently healthy persons. Clin Chem Lab Med 2004;42: [22] Masson S, Latini R, Anand IS, et al. Direct comparison of B-type natriuretic peptide (BNP) and amino-terminal pro BNP in a large population of patients with chronic and symptomatic heart failure: the Valsartan Heart Failure (Val-HeFT) data. Clin Chem 2006;52: [23] Hülsmann M, Berger R, Sturm B, et al. Prediction of outcome by neurohumoral activation, the six-minute walk test and the Minnesota Living with Heart Failure Questionnaire in an outpatient cohort with congestive heart failure. Eur Heart J 2002;23: [24] Richards AM, Doughty R, Nicholls MG, et al. Plasma N-terminal probrain natriuretic peptide and adrenomedullin: prognostic utility and prediction of benefit from carvedilol in chronic ischemic left ventricular dysfunction. Australia New Zealand Heart Failure Group. J Am Coll Cardiol 2001;37: [25] Rothenburger M, Wichter T, Schmid C, et al. Aminoterminal pro type B natriuretic peptide as a predictive and prognostic marker in patients with chronic heart failure. J Heart Lung Transplant 2004;23: [26] Hartmann F, Packer M, Coats AJ, et al. Prognostic impact of plasma N-terminal pro-brain natriuretic peptide in severe chronic congestive heart failure: a substudy of the Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS) trial. Circulation 2004;110: [27] Gardner RS, Chong KS, Morton JJ, McDonagh TA. N-terminal brain natriuretic peptide, but not anemia, is a powerful predictor of mortality in advanced heart failure. J Card Fail 2005;11:S [28] Gardner RS, Chong V, Morton I, McDonagh TA. N-terminal brain natriuretic peptide is a more powerful predictor of mortality than endothelin-1, adrenomedullin and tumour necrosis factor-alpha in patients referred for consideration of cardiac transplantation. Eur J Heart Fail 2005;7: [29] Troughton RW, Prior DL, Pereira JJ, et al. Plasma B-type natriuretic peptide levels in systolic heart failure: importance of left ventricular diastolic function and right ventricular systolic function. J Am Coll Cardiol 2004;43: [30] Richards M, Nicholls MG, Espiner EA, et al. Comparison of B-type natriuretic peptides for assessment of cardiac function and prognosis in stable ischemic heart disease. J Am Coll Cardiol 2006;47: [31] Shelton RJ, Clark AL, Goode K, Rigby AS, Cleland JG. The diagnostic utility of N-terminal pro-b-type natriuretic peptide for the detection of major structural heart disease in patients with atrial fibrillation. Eur Heart J 2006;27: [32] Maeda K, Tsutamoto T, Wada A, et al. High levels of plasma brain natriuretic peptide and interleukin-6 after optimized treatment for heart failure are independent risk factors for morbidity and mortality in patients with congestive heart failure. J Am Coll Cardiol 2000;36: [33] O'Brien RJ, Squire IB, Demme B, Davies JE, Ng LL. Pre-discharge, but not admission, levels of NT-pro BNP predict adverse prognosis following acute LVF. Eur J Heart Fail 2003;5: [34] Cheng V, Kazanagra R, Garcia A, et al. A rapid bedside test for B-type peptide predicts treatment outcomes in patients admitted for decompensated heart failure: a pilot study. J Am Coll Cardiol 2001;37: [35] Bettencourt P, Azevedo A, Pimenta J, Friões F, Ferreira S, Ferreira A. N-terminal-pro-brain natriuretic peptide predicts outcome after hospital discharge in heart failure patients. Circulation 2004;110: [36] Wijeysundera HC, Hansen MS, Stanton E, et al. Neurohormones and oxidative stress in non-ischemic cardiomyopathy: Relationship to survival and the effect of treatment with amlodipine. Am Heart J 2003;146: [37] Bettencourt P, Friões F, Azevedo A, et al. Prognostic information provided by serial measurements of brain natriuretic peptide in heart failure. Int J Cardiol 2004;93:45 8. [38] Anand IS, Fisher LD, Chiang YT, et al. Changes in brain natriuretic peptide and norepinephrine over time and mortality and morbidity in the Valsartan Heart Failure Trial (Val-HeFT). Circulation 2003;107: [39] Latini R, Masson S, Wong M, et al. Incremental prognostic value of changes in B-type natriuretic peptide in heart failure. Am J Med 2006;119(70):e [40] Lainchbury JG, Troughton RW, Frampton CM, et al. NT-pro BNPguided drug treatment for chronic heart failure: design and methods in the BATTLESCARRED trial. Eur J Heart Fail 2006;8: [41] Cleland JGF, Ray S, Dalle Mulle J, et al. Can natriuretic peptides be used to monitor response to carvedilol in patents with ischaemic left ventricular systolic dysfunction? Circulation 2006;106(Suppl II) [Abstract 3487].