Winfried März, 1* Beate Tiran, 2 Ursula Seelhorst, 3 Britta Wellnitz, 3 Johann Bauersachs, 4 Bernhard R. Winkelmann, 5 and Bernhard O.

Transcription

1 Clinical Chemistry 53: (2007) Proteomics and Protein Markers N-Terminal Pro-B-Type Natriuretic Peptide Predicts Total and Cardiovascular Mortality in Individuals with or without Stable Coronary Artery Disease: The Ludwigshafen Risk and Cardiovascular Health Study Winfried März, 1* Beate Tiran, 2 Ursula Seelhorst, 3 Britta Wellnitz, 3 Johann Bauersachs, 4 Bernhard R. Winkelmann, 5 and Bernhard O. Boehm 6 Background: Measurement of N-terminal pro-b-type natriuretic peptide (NT-pro-BNP) measurement can be used to predict mortality in patients with acute coronary syndromes. Information on the value of NT-pro-BNP in clinically stable persons scheduled for angiography is limited. Methods: We used Cox proportional hazards regression to examine the effect of NT-pro-BNP on total and cardiovascular mortality in 1135 with and 506 individuals without stable coronary artery disease (CAD). Results: NT-pro-BNP was associated with New York Heart Association functional class, left ventricular (LV) systolic function, and LV end-diastolic pressure. NTpro-BNP was positively related to age, female sex, hypertension, and former and current smoking and negatively related to body mass index and glomerular filtration rate. During a median follow-up of 5.45 years, NT-pro-BNP concentrations of , , or >2000 ng/l resulted in unadjusted hazard ratios (95% CI) for all-cause death of 3.2 ( ), 6.63 ( ), and 1 Synlab Center of Laboratory Diagnostics, Heidelberg, Germany. 2 Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz, Graz, Austria. 3 Ludwigshafen Risk and Cardiovascular Health Study ggmbh, Freiburg, Germany. 4 Division of Cardiology, Department of Medicine, University of Würzburg, Würzburg, Germany. 5 Cardiology Group, Frankfurt-Sachsenhausen, Germany. 6 Division of Endocrinology, Department of Medicine, University Hospital, Ulm, Germany. * Address correspondence to this author at: Synlab Center of Laboratory Diagnostics Heidelberg, PO Box , D Heidelberg, Germany. Fax ; maerz@synlab.de. Received July 10, 2006; accepted March 14, Previously published online at DOI: /clinchem ( ), respectively, compared with concentrations <100 ng/l. Hazard ratios (CI) for death from cardiovascular causes were 3.8 ( ), 9. 3 ( ), and 22.2 ( ). NT-pro-BNP remained predictive of total and cardiovascular mortality after accounting for age, sex, diabetes mellitus, body mass index, smoking, hypertension, dyslipidemia, glomerular filtration rate, presence or absence of CAD on angiography, cardiovascular medication, revascularization at baseline, clinical signs of heart failure, LV systolic function, and C-reactive protein. Conclusions: NT-pro-BNP is predictive of all-cause and cardiovascular mortality in individuals with or without stable angiographic CAD independently of other cardiovascular risk factors, coronary atherosclerosis, and cardiac function American Association for Clinical Chemistry B-type natriuretic peptide (BNP) 7 is released by the cardiac ventricles in response to increased wall tension. The major biological actions of BNP include vasodilation and increased renal output of sodium and water. BNP is synthesized as a prohormone that is cleaved into active BNP and an inactive N-terminal fragment (NT-pro-BNP) (1). Both BNP and NT-pro-BNP are well-established biomarkers used for diagnosis of heart failure (2). Both BNP and NT-pro-BNP have been shown to predict mortality in 7 Nonstandard abbreviations: BNP, B-type natriuretic peptide; NT-pro- BNP, N-terminal pro-b-type natriuretic peptide; CAD, coronary artery disease; LV, left ventricular; LURIC, Ludwigshafen Risk and Cardiovascular Health; LVEF, LV ejection fraction; CRP, C-reactive protein; GFR, glomerular filtration rate; NYHA, New York Heart Association. 1075

2 1076 März et al.: NT-pro-BNP Predicts Total and Cardiovascular Mortality patients presenting with acute coronary syndromes (3 7) and in asymptomatic individuals in the general population (8, 9). Fewer studies have investigated whether BNP or NTpro-BNP provides prognostic information in patients with stable coronary artery disease (CAD) (10 13) and how this information relates to the clinical assessment of left ventricular (LV) function. We investigated the importance of NT-pro-BNP as a prognostic factor for long-term mortality in clinically stable individuals scheduled for elective coronary angiography. We also assessed whether NT-pro-BNP was related to mortality in individuals without LV systolic dysfunction or clinical signs of heart failure. Materials and Methods study design and participants To examine the relationships between NT-pro-BNP and mortality, we studied participants of the Ludwigshafen Risk and Cardiovascular Health (LURIC) Study. LURIC is an ongoing prospective study of environmental, biochemical, and genetic risk factors for CAD in a hospital-based cohort of white individuals (14). Between June 1997 and January 2000, 3297 German patients who had undergone coronary angiography at the Ludwigshafen General Hospital and did not suffer from major noncardiovascular disease were included. The study was approved by the institutional review board at the Ärztekammer Rheinland-Pfalz. Informed written consent was obtained from each of the participants. Individuals were classified as having diabetes mellitus if plasma glucose was 1.25 g/l in the fasting state or 2.00 g/l 2 h after the oral glucose load (performed in individuals without previous diabetes mellitus diagnosis) (15), or if individuals were receiving oral antidiabetics or insulin. Hypertension was diagnosed if the systolic and/or diastolic blood pressure exceeded 140 and/or 90 mmhg or if there was a significant history of hypertension, mainly evident by the use of antihypertensive medication. Dyslipidemia was defined by categories proposed by the National Cholesterol Education Program Adult Treatment Panel III (16) as HDL cholesterol (HDL-C) 1.0 mmol/l (40 mg/dl) and/or LDL cholesterol (LDL-C) 4.1 mmol/l (160 mg/dl), and/or triglycerides 2.4 (200 mg/dl). CAD was assessed by angiography using the maximum luminal narrowing estimated by visual analysis. In LURIC, clinically relevant CAD was defined as the occurrence of at least 1 stenosis 20% in at least 1 of 15 coronary segments (17). Individuals with stenoses 20% were considered as controls. To examine the impact of other definitions of CAD on the current analysis, we provisionally used the presence of 1 stenosis 50% as a criterion. Furthermore, we stratified our cohort into individuals according to the number of vessels (none, 1, 2, or 3) with stenoses 50%. Of 3279 individuals with coronary angiograms, 1638 presented with unstable angina, non ST-elevation myocardial infarction (troponin T 0.1 g/l, which was the diagnostic cutoff at the time of recruitment of our study population), ST-elevation myocardial infarction (troponin T 0.1 g/l), or valve disease. These patients were excluded from the current evaluation. Thus, we finally included 1641 individuals. In these patients, LV function was graded semiquantitatively by contrast ventriculography into normal or minimally, moderately, or severely impaired. LV end diastolic pressure was available for 1488 patients; it was considered increased at 12 mmhg. Among the 1641 individuals included in the analysis, LV ejection fraction (LVEF), calculated from the right anterior oblique view, was available in 658 patients (40.1%). In this subgroup, the semiquantitative assessment of LV function correlated well with the LVEF (Spearman correlation coefficient 0.696, P 0.001). Information on vital status was obtained from local person registries. The median time of follow-up was 5.45 years. No patients were lost to follow-up. Among the 1641 patients included in this examination, 187 deaths (11.4%) had occurred. A death certificate was obtained for 186 of the decedents and was missing for 1 decedent, who was included in the total mortality analysis but excluded from the cardiovascular mortality analysis. Cardiovascular death included the following categories: sudden death, fatal myocardial infarction, death due to congestive heart failure, death immediately after intervention to treat CAD, fatal stroke, and other causes of death due to CAD. laboratory procedures The standard laboratory methods used have previously been described in detail (14). Lipoproteins were separated by a combined ultracentrifugation-precipitation method. NT-pro-BNP was measured by electrochemiluminescence on an Elecsys 2010 (Roche Diagnostics). Interassay CVs were 3.2% and 2.0% at mean values of 157 and 5125 ng/l. Sensitive C-reactive protein (CRP) was measured by immunonephelometry on a Behring Nephelometer II (N High Sensitivity CRP, Dade Behring). In this assay, the limit of detection for CRP is 0.17 mg/l; it is linear up to 500 mg/l. Glomerular filtration rate (GFR) was calculated as GFR [ml min 1 (1.73 m 2 ) 1 ] 186 creatinine age and GFR (ml min 1 [1.73 m 2 ) 1 ] 138 creatinine age in males and in females, respectively (18). statistical analysis NT-pro-BNP and CRP were transformed logarithmically before being used in parametric statistical procedures. Study participants were divided into 4 groups with concentrations of NT-pro-BNP 100, , , or 2000 ng/l. Clinical and anthropometric characteristics of individuals grouped according to NT-pro-BNP are presented as percentages for categorical variables and as means and SD for continuous variables. Associations of

3 Clinical Chemistry 53, No. 6, categorical and continuous variables with strata of NTpro-BNP were analyzed by 2 test and ANOVA with covariables as indicated, respectively (Table 1). We studied the effect of the angiographic CAD status, sex, age, and risk factors (body mass index, diabetes mellitus, hypertension, smoking, dyslipidemia, and GFR) on NT-pro-BNP with an ANOVA model in which we included those factors not under examination and semiquantitative LV function as covariables (Table 2). To examine the effect of NT-pro-BNP on mortality from all causes and from cardiovascular causes, we calculated hazard ratios and 95% CIs by the Cox proportional hazards regression model. As indicated by log minus log diagnostic plots, the proportional hazards assumption was met throughout. Multivariable adjustment was carried out for sex, age, cardiovascular risk factors (body mass index, diabetes mellitus, hypertension, smoking, dyslipidemia, and GFR), angiographic CAD, previous myocardial infarction, use of cardiovascular drugs (beta-blockers, angiotensin converting enzyme inhibitors, angiotensin type 1 receptor antagonists, calcium channel blockers, diuretics, antiplatelet drugs, and/or lipid lowering agents), revascularization at baseline, and semiquantitative LV function (Tables 3 and 4). All statistical tests were 2-sided. P 0.05 was considered statistically significant. The SPSS 11.0 statistical package (SPSS, SA) was used for all analyses. Results clinical characteristics according to nt-pro-bnp Individuals with high NT-pro-BNP concentrations were, on average, older than those with low concentrations. In this unadjusted analysis, NT-pro-BNP was associated with augmenting prevalence rates of diabetes mellitus, dyslipidemia, systemic hypertension, past smoking, the prevalence and severity of angiographic CAD, previous myocardial infarction, peripheral vascular disease, and Table 1. Clinical and biochemical characteristics associated with NT-pro-BNP. NT-pro-BNP, ng/l < >2000 P a (n 498) (n 641) (n 389) (n 112) Age, years, mean (SD) 56 (10) 62 (10) 65 (10) 65 (12) Male sex, % b b Body mass index, kg/m 2, mean (SD) 28 (4) 28 (4) 27 (4) 26 (4) Diabetes mellitus, % Dyslipidemia, % Systemic hypertension, % Current smoking, % Systolic blood pressure, mmhg mean (SD) 138 (21) 144 (22) b 147 (26) b 140 (25) c Glomerular filtration rate, ml/min 90 (15) 85 (17) 80 (18) 71 (25) CRP, mg/l d Median, 25th and 75th percentile ( ) ( ) ( ) ( ) Angiographic CAD ( 20 percent stenosis), % Angiographic CAD ( 50 stenosis), % Percutaneous intervention at baseline, % b 12.3 b Previous myocardial infarction, % Peripheral vascular disease, % Cerebrovascular disease, % Cardiovascular medication Beta-blockers, % ACE inhibitors, % AT1 receptor antagonists, % Calcium channel blockers, % Diuretics, % Aspirin and/or other antiplatelet agents, % Lipid lowering drugs, % NYHA functional class IV, % LV function severely impaired, % LVEDP 12 mmhg, % LVEF 40, % a ANOVA, adjusted for age and sex, or 2 test. b Significant vs NT-pro-BNP 100 ng/l. c Adjusted for use of beta-blockers, ACE inhibitors, AT1 receptor antagonists, calcium channel blockers, and diuretics. d ANOVA of logarithmically transformed values, adjusted for age and sex.

4 1078 März et al.: NT-pro-BNP Predicts Total and Cardiovascular Mortality Table 2. Effect of cardiovascular risk factors and angiographic status on NT-pro-BNP. n NT-pro-BNP, ng/l a P Men (1.0) Women (1.1) Age, years (1.0) (1.0) (1.1) BMI, kg/m 2 26 or 27 b (1.0) 26 or 27 b (1.0) Diabetes No (1.0) Yes (1.1) Hypertension No (1.1) Yes (1.0) Smoking Never (1.0) Former (1.0) Current (1.1) Dyslipidemia No (1.0) Yes (1.0) GFR, ml/min (1.4) (1.1) (1.0) (1.0) CAD, 20 stenosis No (1.1) Yes (1.0) a Estimated marginal means (SE) obtained in a general linear model (ANOVA), adjusted for each of the other factors and, in addition, semiquantitative LV function, whereby body mass index and age were used as continuous rather than categorial covariables, geometric means are reported for NT-pro-BNP. b Thresholds of 26 and 27 kg/m 2 apply to males and females, respectively. cerebrovascular disease (Table 1). The proportion of men with NT-pro-BNP concentrations between 100 and 399 ng/l was significantly lower than the proportion with concentrations 100 ng/l; it was, however, significantly higher in the small group with NT-pro-BNP 2000 ng/l (P and P by logistic regression, respectively). The rates of percutaneous coronary interventions were higher at NT-pro-BNP concentrations between 100 and 1999 ng/l than concentrations outside this range of values. Systolic blood pressure was significantly higher in patients with NT-pro-BNP between 100 and 1999 ng/l than in those with NT-pro-BNP 100 ng/l. In the subgroup with the highest NT-pro-BNP ( 2000 ng/l), mean systolic blood pressure approximately came down to the mean value seen in the subgroup with the lowest NT-pro- BNP ( 100 ng/l). There was no association of NT-pro- BNP with diastolic blood pressure. Body mass index and GFR decreased as NT-pro-BNP increased. The use of all cardiovascular medications was more prevalent at NTpro-BNP 100 ng/l. As expected, NT-pro-BNP was associated with the New York Heart Association (NYHA) functional class, semiquantitative LV function, the likelihood of increased LV end-diastolic pressure, and the LVEF. association of nt-pro-bnp with cardiovascular risk factors and cad status We examined in more detail the effect of sex, age, risk factors (body mass index, diabetes mellitus, hypertension, smoking), and angiographic status on NT-pro-BNP in a general linear model in which we included those factors not under examination as covariables. We additionally adjusted for semiquantitative LV function. NT-pro-BNP was significantly influenced by sex, age, body mass index, hypertension, smoking status, and GFR. Patients with angiographic CAD had, on average, higher NT-pro-BNP than those without CAD. Diabetes mellitus and dyslipidemia had no significant effect on NT-pro-BNP. nt-pro-bnp and mortality from all causes Among the 1641 patients included in this examination, 187 deaths (11.4%) had occurred during a median observation time of 5.45 years. Compared with individuals with NT-pro-BNP 100 ng/l, the unadjusted hazard ratios for death of patients with NT-pro-BNP of ng/l, ng/l, and 2000 were 3.16 (P 0.001), 6.65 (P 0.001), and (P 0.001), respectively (Table 3, Model 1, Fig. 1). Inclusion of age and sex as covariables decreased these estimates to 2.59 (P 0.001), 4.55 (P 0.001), and (P 0.001), respectively (Table 3, Model 2). Although hazard ratios progressively decreased, NT-pro-BNP retained sizeable prognostic importance after adjustment for cardiovascular risk factors, the presence or absence of CAD, and semiquantitative LV function (Table 3, Model 3). Accounting in addition for CRP did not materially change hazard ratios (Table 3, Model 4). We further analyzed the prognostic value of NT-pro- BNP according to quartiles at baseline. Interquartile ranges were 5 81, , , and ng/l, respectively. Compared with the lowest quartile, crude hazard ratios for death at NT-pro-BNP concentrations in the 2nd, 3rd, and 4th quartiles were 1.73 (P 0.102), 3.36 (P 0.001), and 8.69 (P 0.001), respectively (Table 3, Model 1). As expected, these estimates decreased stepwise with inclusion in the model, of age, sex (Table 3, Model 2), cardiovascular risk factors, angiographic status, and semiquantitative LV function (Table 3, Model 3). Finally, considering CRP as a covariable had no appreciable influence. To study whether NT-pro-BNP added prognostic information beyond clinical assessment, we performed multivariate Cox regression in 1211 individuals in whom LV function was graded normal. Among these, 93 (7.7%) had deceased during the follow-up period. NT-pro-BNP remained significantly predictive of total mortality; ad-

5 Clinical Chemistry 53, No. 6, Table 3. Hazard ratios for mortality from all causes according to NT-pro-BNP. NT-pro-BNP, ng/l, hazard ratio (95% CI) a Deaths, n (%) Model 1 Model 2 Model 3 Model 4 All subjects (n 1641) 100 (n 498) 15 (3.0) (n 642) 59 (9.2) 3.16 ( ) b 2.59 ( ) b 3.18 ( ) c 3.13 ( ) c (n 389) 71 (18.3) 6.63 ( ) b 4.54 ( ) b 5.07 ( ) b 4.96 ( ) b 2000 (n 112) 42 (37.5) ( ) b ( ) b 8.11 ( ) b 7.83 ( ) b 5 81 (n 410) 14 (3.4) (n 411) 24 (5.8) 1.73 ( ) 1.53 ( ) 1.55 ( ) 1.52 ( ) (n 411) 45 (10.9) 3.35 ( ) b 2.55 ( ) d 2.56 ( ) d 2.52 ( ) d (n 409) 104 (25.4) 8.69 ( ) b 5.60 ( ) b 3.80 ( ) b 3.64 ( ) b LV function normal (n 1211) 100 (n 465) 13 (2.8) (n 517) 37 (7.2) 2.62 ( ) d 1.88 ( ) 2.44 ( ) e 2.44 ( ) e 400 (n 229) 43 (18.8) 7.50 ( ) b 4.06 ( ) b 5.15 ( ) e,b 5.12 ( ) b,e LV function normal and NYHA class I (n 602) 100 (n 258) 6 (2.3) (n 258) 12 (4.7) 1.99 ( ) 1.66 ( ) 1.60 ( ) e 1.48 ( ) e 400 (n 86) 15 (17.4) 8.12 ( ) b 5.30 ( ) d ( ) b,e 9.79 ( ) b,e No angiographic CAD (n 506) 100 (n 218) 5 (2.3) (n 190) 9 (4.7) 2.12 ( ) 1.90 ( ) 1.38 ( ) f 1.35 ( ) f 400 (n 98) 13 (13.3) 6.24 ( ) b 4.24 ( ) c 1.84 ( ) f 1.88 ( ) f Angiographic CAD (n 1135) 100 (n 280) 10 (3.6) (n 452) 50 (11.1) 3.23 ( ) b 2.76 ( ) d 3.61 ( ) b,f 3.53 ( ) b,f 400 (n 403) 100 (24.8) 8.02 ( ) b 5.97 ( ) b 6.47 ( ) b,f 6.20 ( ) b,f a Model 1, unadjusted. Model 2, adjusted for age and sex. Model 3, in addition, adjusted for type 2 diabetes, body mass index, smoking status, hypertension, dyslipidemia, GFR, presence or absence of CAD on angiography, previous myocardial infarction, use of beta-blockers, angiotensin converting enzyme inhibitors, angiotensin type 1 receptor antagonists, calcium channel blockers, diuretics, antiplatelet drugs, lipid lowering agents, revascularization at baseline, and semiquantitative LV function. Model 4, in addition, adjusted for CRP. b P c P d P e Semiquantitative LV function not included as a covariable. f Absence or presence of angiographic CAD not included as a covariable. justed and unadjusted hazard ratios were lower than those for the entire study population. We also separately investigated study participants with normal LV function, also classified as NYHA stage I(n 602). Once more, NT-pro-BNP 400 ng/l turned out predictive of all-cause mortality in each of the statistical models. Separate analysis of patients with or without angiographic CAD did not substantially alter our findings (Table 3). In patients with angiographic CAD, NT-pro- BNP remained associated with mortality from all causes in the unadjusted model and in all adjusted models. In individuals without angiographic CAD, the association was significant in the unadjusted model and in the model adjusted for age and sex but turned out insignificant in the models adjusted for risk factors, clinical confounders and CRP (Table 3). Finally, when we used CAD diagnosed as the presence of at least 1 stenosis 50% or when we used the number stenosed of vessels as a covariable in the Cox models instead of CAD diagnosed as at least 1 stenosis 20%, the results were materially unchanged. nt-pro-bnp and mortality from cardiovascular causes Because a death certificate was not available from 1 deceased person, the analysis for cardiovascular mortality included a total of 1640 individuals. Among these, 129 (69.4%) died from cardiovascular causes. Hazard ratios for death from cardiovascular causes according to NTpro-BNP were consistent with those obtained for mortality from all causes in all models (Table 4, Fig. 1), including those that used CAD diagnosed at 1 stenosis 50% or the

6 1080 März et al.: NT-pro-BNP Predicts Total and Cardiovascular Mortality Table 4. Hazard ratios for cardiovascular death according to NT-pro-BNP. NT-pro-BNP, ng/l, hazard ratio (95% CI) a Deaths, n (%) Model 1 Model 2 Model 3 Model 4 All subjects (n 1640) 100 (n 498) 8 (1.6) (n 642) 38 (5.9) 3.82 ( ) b 3.26 ( ) c 2.63 ( ) d 2.59 ( ) d (n 389) 53 (13.6) 9.27 ( ) b 6.78 ( ) b 4.01 ( ) b 3.89 ( ) b 2000 (n 111) 30 (27.0) ( ) b ( ) b 6.21 ( ) b 5.86 ( ) b 5 81 (n 410) 8 (2.0) (n 411) 13 (3.2) 1.64 ( ) 1.49 ( ) 1.35 ( ) 1.33 ( ) (n 411) 30 (7.3) 3.90 ( ) b 3.16 ( ) c 2.35 ( ) d 2.34 ( ) d (n 408) 78 (19.1) ( ) b 7.98 ( ) b 4.05 ( ) b 3.92 ( ) b LV function normal (n 1211) 100 (n 465) 7 (1.5) (n 517) 21 (4.1) 2.76 ( ) d 2.04 ( ) 1.88 ( ) e 1.89 ( ) e 400 (n 229) 26 (11.4) 8.41 ( ) b 4.79 ( ) b 4.01 ( ) c,e 3.96 ( ) c,e LV function normal and NYHA class I (n 602) 100 (n 258) 3 (1.2) (n 258) 4 (1.6) 1.33 ( ) 1.21 ( ) 1.28 ( ) e 1.27 ( ) e 400 (n 86) 10 (11.6) ( ) b 8.03 ( ) c 8.27 ( ) c,e 8.81 ( ) c,e No angiographic CAD (n 506) 100 (n 218) 1 (0.5) (n 190) 4 (2.1) 4.70 ( ) 4.05 ( ) 2.01 ( ) f 1.99 ( ) f 400 (n 98) 11 (11.2) ( ) c ( ) c 8.96 ( ) f 8.93 ( ) f Angiographic CAD (n 1134) 100 (n 280) 7 (2.5) (n 452) 34 (7.5) 3.14 ( ) d 2.80 ( ) d 3.52 ( ) d,f 3.45 ( ) d,f 400 (n 402) 72 (17.9) 8.23 ( ) b 6.57 ( ) b 7.04 ( ) b,f 6.80 ( ) b,f a Model 1, unadjusted. Model 2, adjusted for age and sex. Model 3, in addition, adjusted for type 2 diabetes, body mass index, smoking status, hypertension, dyslipidemia, GFR, presence or absence of CAD on angiography (at least one stenosis 20%, previous myocardial infarction, use of beta-blockers, angiotensin converting enzyme inhibitors, angiotensin type 1 receptor antagonists, calcium channel blockers, diuretics, antiplatelet drugs, lipid lowering agents, revascularization at baseline, and semiquantitative LV function. Model 4, in addition, adjusted for CRP. b P c P d P e Semiquantitative LV function not included as a covariable. f Absence or presence of angiographic CAD not included as a covariable. number of vessels with stenoses 50% as covariables (data not shown). Discussion This study has 2 key findings. First, NT-pro-BNP is a strong predictor of all-cause mortality and cardiovascular mortality in clinically stable patients undergoing coronary angiography, independent of well-established risk factors and the presence of CAD. Second, NT-pro-BNP is also a strong predictor in individuals with no clinical evidence of heart failure or impaired LV function. Abundant information has become available with regard to the utility of BNP and NT-pro-BNP as diagnostic and prognostic markers in heart failure (2, 19, 20). Consistently, we found NT-pro-BNP associated with semiquantitative LV function, NYHA functional class, LVEF, and the likelihood of increased LV end-diastolic pressure. Patients with angiographic CAD had higher average NT-pro-BNP than those without, and there was a positive relationship between the severity of angiographic CAD and NT-pro-BNP. NT-pro-BNP was also influenced by a series of traditional cardiovascular risk factors. As shown previously, after controlling for other confounding factors, women had significantly higher concentrations of NT-pro-BNP than men, and NT-pro-BNP was positively related to age (21). NT-pro-BNP was lower in obese than in lean persons, whereas it was higher in hypertensive than in normotensive persons. Former and current smok-

7 Clinical Chemistry 53, No. 6, Fig. 1. Cumulative survival functions according to NT-pro-BNP in individuals with or without stable angiographic CAD. (A), total mortality; (B), cardiovascular mortality. For hazard ratios and CIs, see Tables 3 and 4. ers had increased NT-pro-BNP concentrations. Similar to BNP, NT-pro-BNP increased as GFR decreased (22). Many studies have dealt with the role of NT-pro-BNP in acute coronary syndromes (3 7), and 4 investigations have addressed the prognostic value of NT-pro-BNP (or BNP) in stable patients undergoing coronary angiography (10 13). In the study by Kragelund et al. (10), the proportions of individuals with LVEF 60% were 80%, 61%, 34%, and 25% as NT-pro-BNP concentrations increased by quartiles. These proportional increments resemble quite closely those in our study participants found to have LV function graded as normal according to NT-pro-BNP concentrations, which were 93%, 81%, 53%, and 21%, respectively, in our 4 groups (Table 1). Furthermore, the distributions of NT-pro-BNP concentrations were similar. Kragelund et al. found interquartile concentration ranges of 5 63, , , and ng/l (10); ours were 5 81, , , and ng/l. Thus, the 2 study populations appear to have had similar LV function, but in our study the atherosclerotic burden seems to have been much lower. When we used the criterion of at least 1 stenosis 20%, the prevalence of angiographic CAD was 69% in our study population; using lumen narrowing of 70% (or 50% in the left main coronary artery) as the cutoff, Kragelund et al. (10) classified 84% of their study participants as having CAD. To directly compare our results with those of Kragelund et al. (10), who analyzed mortality from all causes but not cardiovascular mortality, we examined quartiles of NT-pro-BNP at baseline and also in relation to all-cause mortality. Basically, we found an identical predictive value of NT-pro-BNP, the crude hazard ratio for death in the highest quartile of NT-pro-BNP being 8.69 compared with 6.1 in their study. We suggest, however, that the use of cutoff values of 100, 400, and 2000 ng/l NT-pro-BNP will enable better differentiation than the use of quartiles, because hazard ratios for death in subgroups defined by these cutoff values were roughly 2-fold those found in groups defined by quartiles (Tables 3 and 4). Schnabel et al. (12) examined the association of quartiles of NT-pro-BNP and BNP with cardiovascular death and nonfatal myocardial infarction in 570 and 334 individuals with stable and unstable CAD, respectively. In both groups, they saw a significant association of NT-pro- BNP with future cardiovascular events only in their highest quartile of NT-pro-BNP. Similar results were reported when the same investigators measured BNP rather than NT-pro-BNP in 1085 patients with stable angina (13). At variance to these studies, we consistently detected a continuous relationship between NT-pro-BNP and mortality, both from any cause and from cardiovascular causes, once NT-pro-BNP exceeded 100 ng/l. Remarkably, the association of NT-pro-BNP with mortality from all causes and from cardiovascular causes remained after we adjusted for cardiovascular risk factors, the presence or absence and severity (number of vessels with stenoses 50%) of angiographic CAD, and semiquantitative LV function. The severity of CAD thus did not affect the prognostic value of NT-pro-BNP. Furthermore, in individuals with normal LV function, and in those with both normal LV function and NYHA functional class I, NT-pro-BNP remained predictive, even in fully adjusted models. NT-pro-BNP, therefore, predicts mortality independently, at least in part, from its role as a marker of LV function. The mechanism underlying this observation remains unknown, however. BNP and NTpro-BNP correlate well with the extent of inducible myocardial ischemia on stress single-photon emission computed tomography (23, 24). BNP concentrations increase after short-term coronary occlusion during coronary angioplasty, but not after coronary angiography (25). To-

8 1082 März et al.: NT-pro-BNP Predicts Total and Cardiovascular Mortality gether, these findings may suggest that these peptides reflect chronically recurring myocardial ischemia, but our observation that NT-pro-BNP also predicted death in persons with coronary stenoses 20% does not support this explanation. It is possible, however, that NT-pro-BNP reflects LV dysfunction not detected by the clinical means used here. The utility of measuring BNP and NT-pro-BNP in patients at high risk of death, namely those with chronic heart failure (2, 19, 20) or acute coronary syndromes (3 7), has been firmly established, but information on the use of these markers in other patient groups is limited. Wang et al. (8) demonstrated that BNP was a predictor of all-cause mortality in asymptomatic participants in the Framingham Heart Study. Only recently, in an older community-based population, NT-pro-BNP was a strong risk biomarker for cardiovascular disease and death but CRP was not (9). The current study is the largest to address the clinical value of NT-pro-BNP in clinically stable individuals undergoing angiography. By and large our results are consistent with those of smaller previous investigations (10 13), all suggesting that the measurement of NT-pro-BNP may be helpful in individuals without clinical indications of heart failure. Furthermore, because NT-pro-BNP may be associated with mortality in individuals without relevant angiographic CAD, we speculate that screening of presymptomatic individuals with cardiovascular risk factors might refine risk stratification in the future. Further cohort studies addressing this issue in populations at low or intermediate cardiovascular risk are clearly warranted. Because natriuretic peptide measurement may not be applicable to screening large populations, our results lend further clarity regarding whether screening would be useful in an at-risk population. Grant/funding support: Roche Diagnostics provided unrestricted financial support and assay reagent sets free of charge but did not assume any other role in the conduct of the study. Financial disclosures: Winfried März, Bernhard R. Winkelmann, and Bernhard Boehm are shareholders of the not-forprofit LURIC Study company (LURIC gemeinnützige GmbH). Acknowledgements: We extend our appreciation to the participants of the LURIC Study; without their collaboration this article would not have been written. The technical assistance of Sabine Kern is gratefully acknowledged. We thank all members of the LURIC study team who were either temporarily or permanently involved in patient recruitment and sample and data handling; the laboratory staff at the Ludwigshafen General Hospital and the Universities of Freiburg and Ulm, Germany; and the German registration offices and local public health departments for their assistance. References 1. Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med 1998;339: Doust JA, Glasziou PP, Pietrzak E, Dobson AJ. A systematic review of the diagnostic accuracy of natriuretic peptides for heart failure. Arch Intern Med 2004;164: Omland T, Persson A, Ng L, O Brien R, Karlsson T, Herlitz J, et al. N-terminal pro-b-type natriuretic peptide and long-term mortality in acute coronary syndromes. Circulation 2002;106: de Lemos JA, Morrow DA, Bentley JH, Omland T, Sabatine MS, McCabe CH, et al. The prognostic value of B-type natriuretic peptide in patients with acute coronary syndromes. N Engl J Med 2001;345: Mega JL, Morrow DA, De Lemos JA, Sabatine MS, Murphy SA, Rifai N, et al. B-type natriuretic peptide at presentation and prognosis in patients with ST-segment elevation myocardial infarction: an EN- TIRE-TIMI-23 substudy. J Am Coll Cardiol 2004;44: James SK, Lindahl B, Siegbahn A, Stridsberg M, Venge P, Armstrong P, et al. N-terminal pro-brain natriuretic peptide and other risk markers for the separate prediction of mortality and subsequent myocardial infarction in patients with unstable coronary artery disease: a Global Utilization of Strategies To Open Occluded Arteries (GUSTO)-IV substudy. Circulation 2003;108: Jernberg T, James S, Lindahl B, Stridsberg M, Venge P, Wallentin L. NT-proBNP in unstable coronary artery disease: experiences from the FAST, GUSTO IV and FRISC II trials. Eur J Heart Fail 2004;6: Wang TJ, Larson MG, Levy D, Benjamin EJ, Leip EP, Omland T, et al. Plasma natriuretic peptide levels and the risk of cardiovascular events and death. N Engl J Med 2004;350: Kistorp C, Raymond I, Pedersen F, Gustafsson F, Faber J, Hildebrandt P. N-terminal pro-brain natriuretic peptide, C-reactive protein, and urinary albumin levels as predictors of mortality and cardiovascular events in older adults. JAMA 2005;293: Kragelund C, Gronning B, Kober L, Hildebrandt P, Steffensen R. N-terminal pro-b-type natriuretic peptide and long-term mortality in stable coronary heart disease. N Engl J Med 2005;352: Omland T, Richards AM, Wergeland R, Vik-Mo H. B-type natriuretic peptide and long-term survival in patients with stable coronary artery disease. Am J Cardiol 2005;95: Schnabel R, Rupprecht HJ, Lackner KJ, Lubos E, Bickel C, Meyer J, et al. Analysis of N-terminal-pro-brain natriuretic peptide and C-reactive protein for risk stratification in stable and unstable coronary artery disease: results from the AtheroGene study. Eur Heart J 2005;26: Schnabel R, Lubos E, Rupprecht HJ, Espinola-Klein C, Bickel C, Lackner KJ, et al. B-type natriuretic peptide and the risk of cardiovascular events and death in patients with stable angina: results from the AtheroGene study. J Am Coll Cardiol 2006;47: Winkelmann BR, März W, Boehm BO, Zotz R, Hager J, Hellstern P, et al. Rationale and design of the LURIC study: a resource for functional genomics, pharmacogenomics and long-term prognosis of cardiovascular disease. Pharmacogenomics 2001;2:S American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2006;29(Suppl 1):S Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation 2002;106: Austen WG, Edwards JE, Frye RL, Gensini GG, Gott VL, Griffith LS, et al. A reporting system on patients evaluated for coronary artery disease: Report of the Ad Hoc Committee for Grading of Coronary Artery Disease, Council on Cardiovascular Surgery, American Heart Association. Circulation 1975;51(Suppl 4):5 40.

9 Clinical Chemistry 53, No. 6, Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation: Modification of Diet in Renal Disease Study Group. Ann Intern Med 1999;130: 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: Gardner RS, Ozalp F, Murday AJ, Robb SD, McDonagh TA. N- terminal pro-brain natriuretic peptide: a new gold standard in predicting mortality in patients with advanced heart failure. Eur Heart J 2003;24: Clerico A, Del Ry S, Maffei S, Prontera C, Emdin M, Giannessi D. The circulating levels of cardiac natriuretic hormones in healthy adults: effects of age and sex. Clin Chem Lab Med 2002;40: Wallen T, Landahl S, Hedner T, Saito Y, Masuda I, Nakao K. Brain natriuretic peptide in an elderly population. J Intern Med 1997; 242: Palumbo B, Siepi D, Lupattelli G, Sinzinger H, Fiorucci G, Anniboletti PF, et al. Usefulness of brain natriuretic peptide levels to discriminate patients with stable angina pectoris without and with electrocardiographic myocardial ischemia and patients with healed myocardial infarction. Am J Cardiol 2004;94: Sabatine MS, Morrow DA, de Lemos JA, Omland T, Desai MY, Tanasijevic M, et al. Acute changes in circulating natriuretic peptide levels in relation to myocardial ischemia. J Am Coll Cardiol 2004;44: Tateishi J, Masutani M, Ohyanagi M, Iwasaki T. Transient increase in plasma brain (B-type) natriuretic peptide after percutaneous transluminal coronary angioplasty. Clin Cardiol 2000;23: