Protein Markers. Clinical Chemistry 49: (2003) Proteomics and

Transcription

1 Clinical Chemistry 49: (2003) Proteomics and Protein Markers Prognostic Value of Combination of Cardiac Troponin T and B-Type Natriuretic Peptide after Initiation of Treatment in Patients with Chronic Heart Failure Junnichi Ishii, 1* Wei Cui, 2 Fumihiko Kitagawa, 3 Takahiro Kuno, 3 Yuu Nakamura, 2 Hiroyuki Naruse, 2 Yoshihisa Mori, 2 Takashi Ishikawa, 3 Youichi Nagamura, 4 Takeshi Kondo, 2 Hisaji Oshima, 3 Masanori Nomura, 2 Kouji Ezaki, 2 and Hitoshi Hishida 2 Background: Recent studies have suggested that cardiac troponin T (ctnt) and troponin I may detect ongoing myocardial damage involved in the progression of chronic heart failure (CHF). This study was prospectively designed to examine whether the combination of ctnt, a marker for ongoing myocardial damage, and B-type natriuretic peptide (BNP), a marker for left ventricular overload, would effectively stratify patients with CHF after initiation of treatment. Methods: We measured serum ctnt, plasma BNP, and left ventricular ejection fraction (LVEF) on admission for worsening CHF [New York Heart Association (NYHA) functional class III to IV] and 2 months after initiation of treatment to stabilize CHF (n 100; mean age, 68 years). Results: Mean (SD) concentrations of ctnt [0.023 (0.066) vs (0.20) g/l] and BNP [249 (276) vs 753 (598) ng/l], percentage increased ctnt (>0.01 g/l; 35% vs 60%), NYHA functional class [2.5 (0.6) vs 3.5 (5)], and LVEF [43 (13)% vs 36 (12)%] were significantly (P <0.01) improved 2 months after treatment compared with admission. During a mean follow-up of 391 days, there 1 Division of Critical Care, Fujita Health University Graduate School of Health Sciences, 2 Department of Internal Medicine, Fujita Health University School of Medicine, 3 Department of Joint Research Laboratory of Clinical Medicine, Fujita Health University Hospital, and 4 Department of Clinical Chemistry, Fujita Health University School of Health Sciences, Toyoake , Japan. *Address correspondence to this author at: Division of Critical Care, Fujita Health University Graduate School of Health Sciences, 1-98 Dengakugakubo, Kutsukake-cho, Toyoake , Japan. Fax ; jishii@fujita-hu.ac.jp. Received May 2, 2003; accepted August 28, DOI: /clinchem were 44 cardiac events, including 12 cardiac deaths and 32 readmissions for worsening CHF. On a stepwise Cox regression analysis, increased ctnt and BNP were independent predictors of cardiac events (P <0.001). ctnt >0.01 g/l and/or BNP >160 ng/l 2 months after initiation of treatment were associated with increased cardiac mortality and morbidity rates. Conclusion: The combination of ctnt and BNP measurements after initiation of treatment may be highly effective for risk stratification in patients with CHF American Association for Clinical Chemistry Serum concentrations of cardiac troponin T (ctnt) 5 and cardiac troponin I have been shown to be increased in patients with chronic heart failure (CHF) and to correlate with the severity of the disease and prognosis (1 7). These findings, coupled with the high cardiospecificity of both troponins (8, 9), suggest that ongoing myocardial damage plays an important role in the pathophysiology of CHF and can be documented by increases in cardiac troponins (1 7). The combination of ctnt, a marker for ongoing myocardial damage (1 4), and B-type natriuretic peptide (BNP), a marker for left ventricular overload (10, 11), might effectively stratify patients with CHF. Recently we measured ctnt and BNP concentrations on admission in 98 consecutive patients hospitalized for worsening CHF and monitored these individuals for adverse outcome 5 Nonstandard abbreviations: ctnt, cardiac troponin T; CHF, chronic heart failure; BNP, B-type natriuretic peptide; NYHA, New York Heart Association; ACS, acute coronary syndrome; LVEF, left ventricle ejection fraction; IHD, ischemic heart disease; and CI, confidence interval. 2020

2 Clinical Chemistry 49, No. 12, during a mean follow-up period of 451 days after admission. We showed that ctnt g/l and/or BNP 440 ng/l on admission could reliably stratify the patients into low-, intermediate-, and high-risk groups for cardiac mortality and morbidity (12). Accordingly, this study was prospectively designed to determine whether the combination of measuring ctnt and BNP concentrations after additional medical treatment for CHF could stratify patients with CHF. Materials and Methods We studied 100 consecutive patients, who had been admitted to Fujita Health University hospital for worsening CHF [New York Heart Association (NYHA) functional class III to IV] from June 1999 to June 2001, treated with standard oral drugs for 2 months after admission, and met none of the following criteria. Patients who had clinical or electrocardiographic evidence suggestive of acute coronary syndrome (ACS) for 2 months after treatment had been initiated were excluded from this study. Patients who underwent percutaneous coronary intervention or coronary artery bypass graft during the 2 months were excluded. Patients with a history of recent myocardial infarction or coronary revascularization (within 3 months of admission), myocarditis, renal failure (serum creatinine concentration 30 mg/l), or pulmonary disease were also excluded. Cardiologists not directly involved in this study determined whether patients met these exclusion criteria 2 months after admission. Informed consent was obtained from all patients before participation. Eight of the patients had been included in an earlier study (12). study protocol Venous blood samples were obtained for measurements of ctnt and BNP concentrations on admission and 2 months after treatment had been initiated to stabilize each patient s condition. At the same time, two-dimensional echocardiography was performed by experts blinded to the study, and left ventricular ejection fraction (LVEF) was calculated by use of the modified Simpson rule. All patients underwent clinical follow-up for a mean period of 391 days (range, days). Cardiac events, which were judged by researchers blinded to the biochemistry results, were defined as cardiac death (death from worsening CHF, fatal myocardial infarction, or sudden death) and readmission for worsening CHF or myocardial infarction. Myocardial infarction was defined by a combination of two of three characteristics: typical symptoms, time-dependent changes in serum creatine kinase MB activity, and a typical electrocardiographic pattern involving the development of Q waves. The diagnosis of worsening CHF was established by researchers blinded to biochemistry results on the basis of symptoms, physical findings, and evidence of pulmonary congestion on chest radiography. All patients who did not have a cardiac event underwent clinical follow-up for 12 months. Physicians blinded to the results of the biochemical tests independently selected the appropriate therapy and managed the patients according to standard protocols using outcome measures such as improvement in symptoms, physical findings, and evidence of pulmonary congestion on chest radiography (13). Diuretics were given in flexible dosages on the basis of body weight and daily diuresis. Spironolactone was administered as 25 or 50 mg/day. Angiotensin-converting enzyme inhibitor, angiotensin II receptor blocker, and beta-blocker were gradually increased to the maximum dosage possible for 2 months. The maximum dosages were 5 mg/day for enalapril, 5 mg/day for imidapril, 2 mg/day for temocapril, 50 mg/day for losartan, and 20 mg/day for carvedilol. measurements of biochemical markers Blood samples for measuring serum ctnt were centrifuged at 4 C for 15 min at 1000g and stored at 70 C until assayed. A third-generation Enzymun Test Troponin T assay was used, based on a prototype of the new electrochemiluminescence-based Elecsys system (Roche Diagnostics, Tokyo, Japan). The lower limit of detection of the assay for ctnt was 0.01 g/l. The imprecision for ctnt assay was 14% at g/l and 8.7% at g/l. ctnt was not detectable in 100 healthy adult volunteers. Blood samples for measuring plasma BNP were collected in chilled tubes containing EDTA, disodium salt, and aprotinin (500 IU/mL). The plasma was separated by centrifugation at 4 C for 15 min at 1000g and then stored at 70 C until analysis. BNP concentrations were measured by a commercial RIA for human BNP (Shiono RIA BNP assay; Shionogi Co., Ltd.). The lower limit of detection and the upper limit of the reference interval were 2 and 18.4 ng/l, respectively, for the BNP assay. The intraassay and interassay CV for BNP assay were 5.2% and 6.1%, respectively. data analysis Continuous variables were analyzed using the Mann Whitney U-test, Wilcoxon paired sign-rank test, or linear regression analysis, and the data are expressed as mean (SD). Values below the lower detection limit of the assay were defined as zero. Categorical variables were compared by the 2 test. Univariate and stepwise multivariate Cox regression analyses were used to evaluate the prognostic value of variables. Cardiac event-free survival was determined according to the Kaplan Meier method, and comparison of cardiac event-free survival between subgroups was performed using the log-rank test. ROC curves were used to determine the cutoff values of the biochemical markers for predicting cardiac events (14). P values 0.05 were considered significant.

3 2022 Ishii et al.: ctnt and BNP in Heart Failure Table 1. Comparison of demographics and clinical characteristics of 100 patients with CHF who did or did not die from a cardiac cause and did or did not have a cardiac event. All patients (n 100) No cardiac death (n 88) Cardiac death (n 12) No cardiac event (n 56) Cardiac event (n 44) Age, years 68 (11) 67 (10) 73 (10) a 66 (11) 71 (10) a Gender Male 56 (56%) 50 (57%) 6 (50%) 32 (57%) 24 (55%) Female 44 (44%) 38 (43%) 6 (50%) 24 (43%) 20 (45%) Etiology IHD 37 (37%) 32 (36%) 5 (42%) 17 (30%) 20 (45%) Non-IHD 63 (63%) 56 (64%) 7 (58%) 39 (70%) 24 (55%) Admission/2 months 2 months 2 months 2 months 2 months NYHA class 3.5 (0.5)/2.5 (0.6) b 2.4 (0.5) 2.8 (0.6) a 2.3 (0.4) 2.7 (0.6) b LVEF, % 36 (12)/43 (13) b 44 (13) 34 (10) a 47 (12) 38 (13) b Creatinine, mg/l 11.2 (5.2)/12.1 (5.3) a 11.3 (4.6) 16.3 (6.8) b 10.6 (3.9) 13.5 (6.1) b ctnt, g/l (0.20)/0.023 (0.066) b (0.007) (0.047) b (0.015) (0.094) b (Reference values 0.01 g/l) BNP, ng/l 753 (598)/249 (276) b 225 (268) 426 (279) b 141 (158) 386 (329) b (Reference values 18.4 ng/l) Treatment Diuretics 72/91 81 (92%) 10 (83%) 50 (89%) 41 (93%) Spironolactone 34/94 83 (94%) 11 (92%) 53 (95%) 41 (93%) ACE c inhibitors and/or ARBS 70/ (100%) 12 (100%) 56 (100%) 44 (100%) -Blockers 24/54 48 (55%) 6 (50%) 30 (54%) 24 (55%) a,b Versus admission values, no cardiac death, or no cardiac event: a P 0.05; b P c ACE, angiotensin-converting enzyme; ARB, angiotensin II receptor blocker. Results Patient characteristics are shown in Table 1. The severity of CHF on admission was NYHA functional class III in 54 patients and class IV in 46 patients. The severity of CHF 2 months after additional treatment was NYHA class II in 57 patients, class III in 40 patients, and class IV in 3 patients. The etiologies of CHF were ischemic heart disease (IHD) in 37 patients, hypertensive heart disease in 28 patients, aortic and/or mitral valve disease in 18 patients, dilated cardiomyopathy in 14 patients, atrial fibrillation with rapid ventricular response in 2 patients, and congenital heart disease in 1 patient. Follow-up was complete in all study patients. During the follow-up period, there were 2 noncardiac deaths (1 gastric cancer and 1 pneumonia) and 44 (44%) cardiac events, including 12 cardiac deaths and 32 readmissions for worsening CHF. The causes of cardiac death were worsening CHF in 9 patients and sudden death in 3 patients. At 2 months after the start of treatment, ctnt and BNP concentrations, NYHA functional class, and LVEF were significantly better than at the time of admission (Table 1). The percentage of increased ctnt ( 0.01 g/l) was significantly decreased at 2 months compared with on admission (35% vs 60%; P 0.001). At 2 months after treatment, ctnt was detectable in 11 (19%) of 57 patients in NYHA class II and in 24 (56%) of 43 patients in NYHA class III or class IV. There was a significant but weak correlation between ctnt and log BNP concentrations 2 months after initiation of treatment (r 0.24; P 0.05). Patients who died from a cardiac cause were older (P 0.05); had higher concentrations of ctnt, BNP, and creatinine (P 0.01); had a higher NYHA class (P 0.05); and had a lower LVEF (P 0.05) after treatment was established compared with those who did not die from a cardiac cause (Table 1). Patients who had a cardiac event were older (P 0.05); had higher concentrations of ctnt, BNP, and creatinine (P 0.01); had a higher NYHA class (P 0.01); and had a lower LVEF (P 0.01) after establishment of treatment (Table 1). However, the number of patients with IHD and the treatment regimens were similar in the two groups. The areas under the ROC curves for ctnt and BNP after treatment was initiated [0.73; 95% confidence interval (CI), for ctnt; 0.72; 95% CI, for BNP] did not differ significantly from each other (Fig. 1). The optimal values of ctnt and BNP for predicting cardiac events were defined as the concentration with the largest sum of sensitivity plus specificity for each of the curves; for ctnt, the optimal concentration was 0.01 g/l, and for BNP, it was 160 ng/l. The sensitivity, specificity, positive and negative predictive values, and overall accuracy for predicting cardiac death and cardiac events in patients with a serum ctnt 0.01 g/l were 100% (12 of 12) and 61% (27 of 44), 74% (65 of 88) and 86% (48 of 56), 34% (12 of 35) and 77% (27 of 35), 100% (65 of

4 Clinical Chemistry 49, No. 12, Fig. 1. ROC curves for troponin T (A) and BNP (B) 2 months after initiation of treatment in 100 patients with CHF who did and did not suffer a cardiac event. AUC, area under the curve. 65) and 74% (48 of 65), and 77% (77 of 100) and 75% (75 of 100), respectively. The corresponding sensitivity, specificity, positive and negative predictive values, and overall accuracy in patients with a plasma BNP 160 ng/l were 83% (10 of 12) and 73% (32 of 44), 57% (50 of 88) and 71% (40 of 56), 21% (10 of 48) and 67% (32 of 48), 96% (50 of 52) and 77% (40 of 52), and 60% (60 of 100) and 72% (72 of 100), respectively. The univariate and multivariate predictors of cardiac event and cardiac death are summarized in Table 2. On a stepwise Cox regression analysis including age, sex, etiology, NYHA class, LVEF, creatinine, ctnt, and BNP, only ctnt (P ) and BNP (P ) concentrations after treatment were independent predictors of cardiac events. In addition, when ctnt 0.01 g/l and BNP 160 ng/l after treatment were included in a Table 2. Univariate and multivariate predictors of cardiac death and cardiac events in patients 2 months after initiation of treatment for CHF. Variables Hazard ratio 95% CI P Univariate model of cardiac death and cardiac event Cardiac death Age (increase of 10 years) Gender (male 1) NS a Etiology (IHD 1) NS NYHA class LVEF (increase of 10%) Creatinine (increase of 10 mg/l) ctnt (increase of 0.1 g/l) Log BNP (10-fold increase) Cardiac event Age (increase of 10 years) Gender (male 1) NS Etiology (IHD 1) NYHA class LVEF (increase of 10%) Creatinine (increase of 10 mg/l) ctnt (increase of 0.1 g/l) Log BNP (10-fold increase) Multivariate model of cardiac event Multivariate model including ctnt and BNP concentrations ctnt (increase of 0.1 g/l) Log BNP (10-fold increase) LVEF (increase of 10%) Multivariate model including ctnt 0.01 g/l and BNP 160 ng/l as continuous variables ctnt 0.01 g/l BNP 160 ng/l LVEF (increase of 10%) a NS, not significant.

5 2024 Ishii et al.: ctnt and BNP in Heart Failure multivariate model as continuous variables, ctnt 0.01 g/l (P ) and BNP 160 ng/l (P 0.013) were independent predictors of cardiac events. A ctnt concentration 0.01 g/l and/or BNP 160 ng/l 2 months after initiation of treatment correlated with an incremental increase in cardiac mortality and cardiac morbidity rates (Fig. 2). With this combination, patients with or without BNP 160 ng/l could be further classified into two different risk groups for cardiac events (group II vs group IV, P 0.01; group I vs group III, P 0.1) and cardiac death (group II vs group IV, P 0.01; group I vs group III, P 0.05). In addition, patients with or without ctnt 0.01 g/l could be further classified into two different risk groups for cardiac event (group III vs group IV, P 0.1; group I vs group II, P 0.1). Thus the combination of ctnt and BNP concentrations after additional treatment appeared to improve the ability to stratify risk in patients with CHF over the individual use of these markers, especially BNP alone. Kaplan Meier analysis showed that this combination could be used to stratify the patients into the four groups for cardiac events (Fig. 3). Fig. 3. Cardiac event-free survival curves for the four risk groups of patients with CHF, based on ctnt and BNP concentrations 2 months after initiation of treatment. Group I, ctnt 0.01 g/l and BNP 160 ng/l; group II, ctnt 0.01 g/l and BNP 160 ng/l; group III, ctnt 0.01 g/l and BNP 160 ng/l; and group IV, ctnt 0.01 g/l and BNP 160 ng/l. Discussion The present study demonstrates that, similar to BNP concentrations (10, 11, 15, 16), ctnt concentrations after treatment is initiated are independently associated with cardiac events in patients with CHF and that a ctnt 0.01 g/l and/or BNP 160 ng/l after initiation of treatment could stratify patients into four groups for cardiac mortality and morbidity. Thus the combination of measuring ctnt, a marker for ongoing myocardial damage, and BNP, a marker for left ventricular overload, represents a highly effective means of risk stratification after additional medical treatment even in clinically stabilized CHF patients. Serum ctnt after treatment was initiated decreased in parallel with the decrease in BNP concentrations, improvement in LVEF, and symptom reduction, suggesting that ongoing myocardial damage may be partly suppressed by the treatment of CHF. Serum ctnt was still measurable 2 months after initiation of treatment in 35% of clinically stable patients with CHF. Such patients were at increased risk for cardiac death and cardiac events. ctnt was detectable after treatment was begun in all 12 patients who died of a cardiac death, and the ctnt concentration after initiation of treatment was an independent predictor of cardiac morbidity. These findings suggested that ongoing myocardial damage may play an important role in the progression of CHF and lead to an adverse outcome (1 7). In the present study, the combination of ctnt and BNP after treatment was able to stratify these clinically stable patients into four groups for adverse outcomes. The weak correlation between ctnt and BNP concentrations suggests that markers specific for ongoing myocardial dam- Fig. 2. Cardiac mortality (A) and cardiac event rates (B) for the four risk groups of patients with CHF, based on ctnt and BNP concentrations 2 months after initiation of treatment. Group I, ctnt 0.01 g/l and BNP 160 ng/l; group II, ctnt 0.01 g/l and BNP 160 ng/l; group III, ctnt 0.01 g/l and BNP 160 ng/l; group IV, ctnt 0.01 g/l and BNP 160 ng/l.

6 Clinical Chemistry 49, No. 12, age and left ventricular overload reflect different aspects of the pathophysiology of CHF and may identify different groups of patients at risk. It is reasonable to expect that patients who have both ongoing myocardial damage and left ventricular overload are at greatest risk. The authors of a recent study reported that aminoterminal probnp-guided treatment of heart failure reduced total cardiovascular events and delayed time to first event compared with intensive clinically guided treatment (17). In the present study, the combination of ctnt and BNP concentrations after additional treatment appeared to improve the ability to stratify risk in patients with CHF over the use of these markers individually, particularly BNP alone. Thus, our results indicate that the combination of ctnt and BNP might be used to guide the treatment of patients with CHF. When patients are categorized as group II or especially IV after additional treatment, further treatment and intensive care might improve the prognosis. When patients are rated as group III after additional treatment, even in the absence of high concentrations of BNP, further treatment to decrease the ctnt concentrations might be needed to reduce cardiac events. Additional studies are needed before treatment decisions can be based on these measurements. The multivariate analysis showed that LVEF tended to be an independent predictor of cardiac events. These findings indicate that combinations of LVEF and ctnt, LVEF and BNP, or of both biochemical markers together with LVEF might effectively stratify patients with CHF. Further studies are needed to clarify the utility of these combinations in patients with CHF. It is difficult to completely exclude the possibility that increased concentrations of BNP and, particularly, ctnt might be related to ischemic cell injury attributable to ACS. We therefore investigated the association of adverse outcome with these markers 2 months after initiation of treatment, when patients were in stable clinical condition. We also ruled out patients who had clinical or electrocardiographic evidence suggestive of ACS or those who received coronary revascularization for 2 months after the initiation of treatment. We believe that this strategy can minimize the possibility that increased concentrations of these markers after treatment are caused by ischemic cell injury attributable to ACS. Recent studies have shown that increased concentrations of ctnt (18 21) and BNP (22, 23) may be independently associated with poor prognosis in patients undergoing chronic hemodialysis. It would be interesting to determine whether our findings might apply to patients with severe renal failure. However, to minimize the effects of renal clearance, we excluded patients with serum creatinine concentrations 30 mg/l. In the present study, there was a relatively small number of cardiac events (n 44) for every putative predictor (n 8) included in multivariate models. In addition, we could not use multivariate models to determine independent predictors of cardiac mortality because of the low number of cardiac deaths (n 12). Thus, additional studies should be carried out in a larger CHF population to confirm our results. Threshold values derived from the ROC curves are highly dependent on the study population and might have been different in another set of patients. Thus, the threshold values for ctnt and BNP that were used in the present study should be confirmed in larger follow-up studies. In addition, in the present study treatments were not randomized. Thus, it was difficult to evaluate the effects of specific drugs for CHF on adverse outcomes. In conclusion, the combination of ctnt and BNP measurements after initiation of treatment may be highly effective for risk stratification in patients with CHF. We would like to thank Roche Diagnostics for providing the required reagents and instruments. References 1. Missov E, Mair J. A novel biochemical approach to congestive heart failure: cardiac troponin T. Am Heart J 1999;138: Del Carlo CH, O Connor CM. Cardiac troponins in congestive heart failure. Am Heart J 1999;138: Setsuta K, Seino Y, Takahashi N, Ogawa T, Sasaki K, Harada A, et al. Clinical significance of elevated levels of cardiac troponin T in patients with chronic heart failure. Am J Cardiol 1999;84: Setsuta K, Seino Y, Ogawa T, Arao M, Miyatake Y, Takano T. Use of cytosolic and myofibril markers in the detection of ongoing myocardial damage in patients with chronic heart failure. Am J Med 2002;113: Missov E, Calzolari C, Pau B. Circulating cardiac troponin I in severe congestive heart failure. Circulation 1997;96: La Vecchia L, Mezzena G, Ometto R, Finocchi G, Bedogni F, Soffiati G, et al. Detectable serum troponin I in patients with heart failure of nonmyocardial ischemic origin. Am J Cardiol 1997;80: La Vecchia L, Mezzena G, Zanolla L, Paccanaro M, Varotto L, Bonanno C, et al. Cardiac troponin I as diagnostic and prognostic marker in severe heart failure. J Heart Lung Transplant 2000;19: Katus HA, Remppis A, Neumann FJ, Scheffold T, Diederich KW, Vinar G, et al. Diagnostic efficacy of troponin T measurements in acute myocardial infarction. Circulation 1991;83: Adams JE 3rd, Bodor GS, Davila-Roman VG, Delmez JA, Apple FS, Ladenson JH, et al. Cardiac troponin I. A marker with high specificity for cardiac injury. Circulation 1993;88: Tsutamoto T, Wada A, Maeda K, Hisanaga T, Maeda Y, Fukai D, 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: Maeda K, Tsutamoto T, Wada A, Hisanaga T, Kinoshita M. Plasma brain natriuretic peptide as a biochemical marker of high left ventricular end-diastolic pressure in patients with symptomatic left ventricular dysfunction. Am Heart J 1998;135: Ishii J, Nomura M, Nakamura Y, Naruse H, Mori Y, Ishikawa T, et al. Risk stratification using a combination of cardiac troponin T and brain natriuretic peptide in patients hospitalized for worsening chronic heart failure. Am J Cardiol 2002;89: Committee on Evaluation and Management of Heart Failure.

7 2026 Ishii et al.: ctnt and BNP in Heart Failure Guidelines for the evaluation and management of heart failure. Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 1995; 92: Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982; 143: Maeda K, Tsutamoto T, Wada A, Mabuchi N, Hayashi M, Tsutsui T, 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: Richards AM, Doughty R, Nicholls MG, MacMahon S, Ikram H, Sharpe N, et al. Neurohumoral prediction of benefit from carvedilol in ischemic left ventricular dysfunction. Circulation 1999;99: 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: Dierkes J, Domrose U, Westphal S, Ambrosch A, Bosselmann HP, Neumann KH, et al. Cardiac troponin T predicts mortality in patients with end stage renal disease. Circulation 2000;102: Ooi DS, Zimmerman D, Graham J, Wells GA. Cardiac troponin T predicts long-term outcomes in hemodialysis patients. Clin Chem 2001;47: Ishii J, Nomura M, Okuma T, Minagawa T, Naruse H, Mori Y, et al. Risk stratification using serum concentrations of cardiac troponin T in patients with end-stage renal disease on chronic maintenance dialysis. Clin Chim Acta 2001;312: Apple FS, Murakami MM, Pearce LA, Herzog CA. Predictive value of cardiac troponin I and T for subsequent death in end-stage renal disease. Circulation 2002;106: Zoccali C, Mallamaci F, Benedetto FA, Tripepi G, Parlongo S, Cataliotti A, et al., for Creed Investigators. Cardiac natriuretic peptides are related to left ventricular mass and function and predict mortality in dialysis patients. J Am Soc Nephrol 2001;12: Naganuma T, Sugimura K, Wada S, Yasumoto R, Sugimura T, Masuda C, et al. The prognostic role of brain natriuretic peptides in hemodialysis patients. Am J Nephrol 2002;22: