The American Journal of Medicine
Volume 122, Issue 3 , Pages 265-272, March 2009

B-Type Natriuretic Peptides for the Evaluation of Exercise Intolerance

  • Christian Mueller, MD

      Affiliations

    • Department of Internal Medicine, University Hospital Basel, Switzerland
    • Corresponding Author InformationRequests for reprints should be addressed to Christian Mueller, MD, FESC, Department of Internal Medicine, University Hospital Basel, Petersgraben 4, CH-4031 Basel, Switzerland
    • ,
  • Micha T. Maeder, MD

      Affiliations

    • Department of Cardiology, University Hospital Basel, Switzerland
    • ,
  • Andreas Christ, MD

      Affiliations

    • Department of Internal Medicine, University Hospital Basel, Switzerland
    • ,
  • Tobias Reichlin, MD

      Affiliations

    • Department of Internal Medicine, University Hospital Basel, Switzerland
    • ,
  • Daniel Staub, MD

      Affiliations

    • Department of Internal Medicine, University Hospital Basel, Switzerland
    • ,
  • Markus Noveanu, MD

      Affiliations

    • Department of Internal Medicine, University Hospital Basel, Switzerland
    • ,
  • Tobias Breidthardt, MD

      Affiliations

    • Department of Internal Medicine, University Hospital Basel, Switzerland
    • ,
  • Mihael Potocki, MD

      Affiliations

    • Department of Internal Medicine, University Hospital Basel, Switzerland
    • ,
  • Martin H. Brutsche, MD, PhD

      Affiliations

    • Division of Pulmonary Care, University Hospital Basel, Switzerland

Article Outline

Abstract 

Background

Cardiopulmonary exercise testing is the method of choice for the differentiation of exercise intolerance. This study sought to assess the utility of B-type natriuretic peptide (BNP) and N-terminal-pro-B-type natriuretic peptide (NT-proBNP) for the identification of a cardiocirculatory exercise limitation.

Methods

In 162 patients undergoing cardiopulmonary exercise testing, rest and peak exercise BNP and NT-proBNP levels were measured. In 94 patients fulfilling criteria for appropriate effort and sufficient diagnostic certainty, the accuracy of BNP and NT-proBNP for the prediction of a cardiocirculatory limitation, as assessed based on clinical and exercise testing data, was determined.

Results

A cardiocirculatory limitation was identified in 27 (29%) patients. Median (interquartile range) resting BNP [162 (45-415) vs 39 (19-94) vs 24 (15-46) pg/mL; P <.001] and NT-proBNP [506 (129-1167) vs 77 (35-237) vs 34 (19-77) pg/mL; P <.001] were higher in patients with cardiocirculatory as compared with those with pulmonary limitation (n=28) and those without cardiocirculatory or pulmonary limitation (n=39). The area under the receiver operator characteristics curve for BNP and NT-proBNP to identify a cardiocirculatory limitation was 0.79 and 0.84, respectively (P=.15 for comparison of the curves). Sensitivity and specificity of the optimal BNP cutoff of 85 pg/mL were 63% and 84%, respectively. Sensitivity and specificity of the optimal NT-proBNP cutoff of 223 pg/mL were 74% and 85%, respectively. Peak exercise biomarkers were not more accurate than resting levels.

Conclusions

Among patients referred for cardiopulmonary exercise testing for evaluation of unexplained exercise intolerance, BNP and NT-proBNP were similarly useful to identify those with a cardiocirculatory limitation.

Keywords: Cardiopulmonary exercise testing, Exercise tolerance, Natriuretic peptides, Sensitivity, Specificity

 

Exercise intolerance is a common symptom, but evaluation of the underlying cause is often challenging. Standard tests performed at rest may reveal abnormalities in both the cardiovascular and the pulmonary system, but cannot reliably predict the exercise response.1 However, identification of the type of exercise limitation (ie, cardiocirculatory vs pulmonary) is mandatory for the initiation of appropriate treatment.

Clinical Significance

 


•Cardiopulmonary exercise testing is the gold standard for the evaluation of exercise intolerance; the role of biomarkers in this setting is unknown.

•B-type natriuretic peptide (BNP) and N-terminal-pro-B-type natriuretic peptide (NT-proBNP) measured at rest had a high accuracy rate for the identification of a cardiocirculatory exercise limitation.

•Measurement of BNP and NT-proBNP at peak exercise did not add to the information obtained from rest levels.

Cardiopulmonary exercise testing provides a global assessment of the exercise response, and is the accepted gold standard for the evaluation of exercise intolerance and unexplained dyspnea.1 Unfortunately, cardiopulmonary exercise testing requires considerable infrastructure and expertise, and is therefore not widely applied. Thus, a simple and easily readable test revealing the patient's exercise limitation with similar accuracy to cardiopulmonary exercise testing would be highly desirable.

The use of biomarkers has emerged as an attractive option in this context. B-type natriuretic peptide (BNP) and the amino-terminal part of its precursor peptide (N-terminal-pro-B-type natriuretic peptide; NT-proBNP) have proved to be helpful in the differentiation of heart failure from other causes of dyspnea in patients presenting to the emergency department with acute dyspnea.2, 3, 4, 5 Similarly, BNP and NT-proBNP also might be helpful for the identification of an exercise limitation due to the cardiovascular system.

BNP and NT-proBNP differ considerably with respect to their plasma half-life (20 minutes vs. 1-2 hours).6 Thus, a head-to-head evaluation of the 2 markers for a new possible application makes sense. In addition, as both BNP and NT-proBNP levels increase in response to exercise,7, 8 we hypothesized that exercise-induced changes in BNP and NT-proBNP might be higher in patients with cardiocirculatory limitation. Thus, the aim of the present study was to compare the accuracy of BNP and NT-proBNP at rest and peak exercise to identify a cardiocirculatory limitation in patients referred for cardiopulmonary exercise testing for evaluation of unexplained exercise intolerance.

Back to Article Outline

Methods 

Patient Population 

From May 2004 to June 2005, 185 consecutive patients referred for cardiopulmonary exercise testing for the evaluation of unexplained exercise intolerance at the University Hospital Basel were eligible. “Exercise intolerance” was defined as the symptom of feeling unable to exercise to a desired level, and it was not synonymous with objectively measured impaired exercise capacity. We studied only patients with chronic symptoms. In accordance with cardiopulmonary exercise testing guidelines,1 any acutely ill patients were excluded from the study. Otherwise, there were no exclusion criteria. Twelve patients did not consent, and in 11 patients, biomarkers were not measured. From the remaining 162 patients undergoing the study procedure, we excluded those in whom, based on criteria outlined below, exercise quality was insufficient due to poor effort (n=22), and patients in whom the cause of exercise limitation could not be determined with sufficient certainty (diagnostic certainty <70% [see section “Test Interpretation and Identification of a Cardiocirculatory Exercise Limitation”], n=46]. This left a total of 94 patients for the present analysis. The study was approved by the local ethics committee.

Cardiopulmonary Exercise Testing 

Patients underwent symptom-limited upright cycle ergometry (Jaeger, Wuerzburg, Germany) using ramp protocols (10-20 watt increase per minute) with continuous electrocardiographic monitoring. Expired gases were acquired continuously, and oxygen uptake (VO2) and carbon dioxide output (VCO2) were recorded in rolling 30-second averages. Calibration of the air flow system was performed before each test with a 3-liter syringe, and calibration of the gas sensors was performed daily. The respiratory exchange ratio (RER), defined as VCO2 divided by VO2, was determined throughout the test. Arterial blood was obtained at rest and peak exercise for blood gas analysis. Exercise quality was judged based on the following criteria: 1) peak heart rate >80% predicted, 2) RER >1.2, 3) lactate at peak exercise >4.0 mmol/L, or 4) difference between base excess at rest and peak exercise <−2.5 mmol/L.9 If none of these criteria was fulfilled, poor effort was assumed, and the patient was excluded from the analysis.

Test Interpretation and Identification of a Cardiocirculatory Exercise Limitation 

An integrative approach as suggested by the American Thorax Society/American College of Chest Physicians statement on cardiopulmonary exercise testing was used.1 The reference standard was defined by a single expert (MHB) using exercise testing data in conjunction with all other information pertaining to the individual patient, including history, physical examination, electrocardiogram, chest radiograph, echocardiogram, pulmonary function tests, and computed tomography scans of the chest. Blinded to BNP and NT-proBNP levels, the expert decided whether a patient was primarily limited by the cardiovascular (cardiocirculatory limitation) or the pulmonary system (either ventilatory limitation or limitation due to limited lung diffusion capacity, pulmonary limitation), or whether neither a cardiocirculatory nor a pulmonary limitation was present (reference group). Patients with hyperventilation were allocated to this latter group. The diagnostic certainty of this assessment was quantified (0-100%). Only patients with a diagnostic certainty ≥70% were included in the analysis.

Test interpretation was based on an algorithm, which has been used for years at our institution, and which has recently been published.9 Indicators for pulmonary limitation included pathologic peak exercise alveolo-arterial oxygen pressure difference [D(A-a)O2] (age-adjusted values), hypoxemia/desaturation during or at peak exercise, and reduced breathing reserve at peak exercise (cutoff 30%), as well as indications of flow limitation on flow volume loops during different exercise levels. In contrast, an inappropriate VO2 increase for a given increase in work rate (ΔVO2/ΔWR), a low oxygen pulse, an impaired chronotropic response despite other evidence of effort, and the presence of ischemic ST-segment changes were regarded as indicators of cardiocirculatory limitation. An increased peak exercise dead space-to-tidal volume ratio (VD/VT) was generally seen as an indicator of pulmonary limitation. However, given that ventilatory inefficiency also is a key feature of heart failure,10 it was not used as an argument against a cardiocirculatory limitation if this was suggested by other features. Ventilatory inefficiency also was assessed by the peak exercise ventilation/VCO2 ratio.11

Blood Sampling and Laboratory Methods 

A specimen of venous blood was drawn before and 1 minute after peak exercise in the seated position from a catheter previously inserted into an antecubital vein. These samples were collected in plastic tubes containing ethylene-diamine-tetra-acetate, placed on ice and centrifuged at 3000 g. BNP concentration was determined using the AxSYM BNP assay (Abbott Laboratories, Zug, Switzerland).12 The coefficients of variation within an assay are 6.0%, 4.3%, and 5.1% for concentrations of 108 pg/mL, 524 pg/mL, and 2117 pg/mL, respectively, and the respective coefficients of variation between assays are 8.1%, 7.5%, and 10%. Plasma levels of NT-proBNP were determined with the Elecsys proBNP assay (Roche Diagnostics, Basel, Switzerland).13 The intra-assay coefficients of variation are 2.4% and 1.8% at 355 pg/mL and 4962 pg/mL, respectively, and the respective interassay coefficients of variation are 2.9% and 2.3%. The laboratory technician performing the assays was at a different site and blinded to patient characteristics and exercise testing data.

Statistical Analysis 

Data were expressed as counts and percentages, mean± standard deviation, or median (interquartile range) as appropriate. Comparisons among the 3 groups were made using analysis of variance, Kruskal-Wallis tests, chi-squared or Fisher's exact tests. BNP and NT-proBNP levels at rest and peak exercise (ln-transformed due to skewed distribution) in the 3 groups were compared using the general linear model for repeated measures. Multivariate logistic regression was performed to identify predictors of cardiocirculatory limitation. Historical data and resting measurements including blood gas analysis were entered as covariates (ln-transformation for covariates with skewed distribution). One model was developed with BNP, and a second model with NT-proBNP as covariate. Receiver operator characteristics curves were constructed to assess sensitivity and specificity of BNP and NT-proBNP at rest and at peak exercise throughout the concentrations to detect cardiocirculatory limitation. A P value <.05 was considered statistically significant. Analysis was performed using commercially available software packages (SPSS/PC, version 15.0, SPSS Inc, Chicago, IL, and Analyse-it V2.04, Leeds, UK).

Back to Article Outline

Results 

Patient Characteristics 

The 68 patients excluded from the analysis were older (61±15 vs 53±15 years; P=.001) and had higher D(A-a)O2 at rest (25±12 vs 20±14 mm Hg; P=.01), higher VD/VT at rest (0.31±0.06 vs 0.29±0.07; P=.04) and peak exercise (0.17±0.07 vs 0.14±0.06; P=.003), lower percent predicted (69±17 vs 69±17%; P=.016) and body weight-indexed peak VO2 (17.0±5.6 vs 21.0±8.2 mL/kg/min; P=.001), and lower lactate at peak exercise (5.6±3.5 vs 6.7±3.1 mmol/L; P=.049), whereas sex, other baseline characteristics, all other measurements at rest and peak exercise, and BNP [44 (21-96) vs 38 (15-140); P=.58] and NT-proBNP [116 (45-249) vs 80 (27-190); P=.41] at rest as well BNP and NT-proBNP concentrations at peak exercise and changes during exercise did not differ between patients excluded from the study and those included in the study.

Among the 94 patients included in the analysis, 39 (41%) had neither a cardiocirculatory nor a pulmonary limitation (reference group), 28 (30%) had pulmonary limitation, and 27 (29%) had cardiocirculatory limitation. In Table 1, patient characteristics are compared across the 3 study groups.

Table 1. Baseline Characteristics
Reference Group (n=39)Pulmonary Limitation (n=28)Cardiocirculatory Limitation (n=27)P Value
Age (years)46±1757±13⁎57±12†.003
Sex (male)18(46%)17(61%)18(67%).22
Body mass index (kg/m2)25.2±4.826.1±4.126.1±5.2.68
Medical history
Coronary artery disease2(5%)4(14%)12(44%)†‡<.001
Other cardiopathy3(5%)1(4%)4(12%).29
Diabetes2(5%)2(7%)6(23%).07
Hypertension7(20%)10(36%)12(42%).06
Chronic obstructive lung disease3(8%)9(32%)⁎6(22%).04
Asthma bronchiale5(13%)2(7%)0.15
Other pneumopathy12(30%)11(39%)7(27%).55
Medical therapy
Beta-blocker4(10%)7(25%)14(52%)†.001
ACEI/ARB8(20%)5(18%)12(46%)‡.04
Diuretic2(5%)5(18%)12(44%)†<.001
Statin5(13%)6(21%)13(48%)†.004
Aspirin3(8%)6(21%)12(44%)†.002
Inhaled β2-mimetics10(26%)8(29%)5(19%).67
Inhaled corticosteroids10(26%)7(25%)2(7%).15
Dyspnea functional class
NYHA I/II/III/IV10(26%)/24(61%)/4(10%)/1(3%)8(29%)/12(42%)/8(29%)/04(15%)/12(44%)/11(41%)/0.11

ACEI=angiotensin-converting-enzyme inhibitor; ARB=angiotensin receptor blocker; NYHA=New York Heart Association.

Data are given as counts and percentages and/or mean±standard deviation.

Post hoc tests

⁎P <.05 reference vs pulmonary limitation;

†P <.05 reference vs cardiocirculatory limitation;

‡P <.05 pulmonary vs cardiocirculatory limitation.

Measurements at Rest 

In Table 2, measurements available at rest are compared among the 3 groups. Left ventricular ejection fraction, creatinine clearance, and the arterial carbon dioxide pressure (PaCO2) were the only parameters distinguishing patients with cardiocirculatory limitation from the other 2 groups.

Table 2. Measurements at Rest
Reference Group (n=39)Pulmonary Limitation (n=28)Cardiocirculatory Limitation (n=27)P Value
Heart rate (beats per minute)79±1280±1481±15.80
Systolic blood pressure (mm Hg)119±16123±14124±27.51
Left ventricular ejection fraction (%)61 (60-65) n=1560 (55-65) n=1542 (25-56)†‡ n=18<.001
Creatinine clearance (mL/min)105±35106±3183±29†‡.02
Hemoglobin (g/dL)14.6±1.414.1±1.713.8±2.1.14
FEV1 (l)2.75±0.781.94±0.81⁎2.56±0.58‡<.001
Oxygen saturation (%)96 (95-97)95 (93-96)95 (92-96)†.01
PaO2 (mm Hg)89.1±11.279.9±10.4⁎81.9±18.1.015
PaCO2 (mm Hg)35.5±5.037.3±6.632.5±5.2†‡.008
Alveolar-arterial oxygen pressure gradient (mm Hg)13±1022±11⁎28±17†<.001
VD/VT0.27±0.080.31±0.060.30±0.06.07

FEV1=forced expiratory volume within the first second; PaO2=arterial oxygen pressure; PaCO2=arterial carbon dioxide pressure; VD/VT=dead space-to-tidal volume ratio.

Data are given as mean±standard deviation or median (interquartile range).

Post hoc tests

⁎P <.05 reference vs pulmonary limitation;

†P <.05 reference vs cardiocirculatory limitation;

‡P <.05 pulmonary vs cardiocirculatory limitation.

Exercise Tests 

Detailed results from cardiopulmonary exercise testing are presented in Table 3. Patients with cardiocirculatory limitation had lower body weight-indexed peak VO2, lower peak heart rate, lower peak systolic blood pressure, and higher breathing reserve than the other 2 groups.

Table 3. Measurements at Peak Exercise
Reference Group (n=39)Pulmonary Limitation (n=28)Cardiocirculatory Limitation (n=27)P Value
Quality of the test
Respiratory exchange ratio1.18±1.131.19±0.141.17±0.15.90
% Predicted heart rate (%)84±1285±974±16†‡.003
Lactate (mmol/L)7.5±3.16.4±2.95.7±3.1.06
Exercise capacity
Exercise time (minutes)6.6 (5.9-10.5)5.8 (5.0-7.8)5.3 (4.2-6.1) †.002
% Predicted peak VO2(%)95±2281±18⁎66±23†‡<.001
Body weight-corrected peak VO2(mL/kg/min)25.9±8.119.9±7.2⁎15.0±4.1†‡<.001
ΔVO2/ΔWR (mL/min/W)10.4±1.99.2±1.68.7±2.2†.002
Exercise response
Heart rate (beats per minute)146±26139±21120±28†‡.001
Systolic blood pressure (mm Hg)183±31188±29156±29†‡<.001
% Predicted oxygen pulse (%)103±2682±12⁎73±23†<.001
Breathing reserve (%)30±1911±20⁎42±14†‡<.001
Oxygen saturation (%)96 (95-97)94 (90-96)⁎95 (91-96)†.005
PaO2 (mm Hg)100.9±11.282.3±19.1⁎87.5±21.8†<.001
PaCO2 (mm Hg)33.6±5.938.0±6.7⁎32.9±4.4‡.003
Alveolar-arterial oxygen pressure gradient (mm Hg)16±1032±16⁎29±22†<.001
VD/VT0.10±0.050.16±0.06⁎0.17±0.05†<.001
VE/VCO232.6 (29.6-36.7)35.3 (32.6-40.8)37.9 (34.6-50.5)†<.001

ΔVO2/ΔWR=increase in oxygen consumption for a given increase in work rate; Peak VO2=peak oxygen consumption; PaO2=arterial oxygen pressure; PaCO2=arterial carbon dioxide pressure; VD/VT=physiological dead space-to-tidal volume ratio; VE/VCO2=relationship between minute ventilation and carbon dioxide production at peak exercise.

Data are given as mean±standard deviation or median (interquartile range).

Post hoc tests

⁎P <.05 reference vs pulmonary limitation;

†P <.05 reference vs cardiocirculatory limitation;

‡P <.05 pulmonary vs cardiocirculatory limitation.

BNP and NT-proBNP Kinetics 

As shown in Table 4, there was an increase in BNP and NT-proBNP from rest to exercise in all 3 groups. BNP and NT-proBNP at rest and peak exercise were higher in patients with cardiocirculatory limitation as compared with patients with pulmonary limitation and the reference group, and NT-proBNP but not BNP at rest and at peak exercise was higher in patients with pulmonary limitation than in the reference group. BNP and NT-proBNP both at rest (r=0.80; P <.001) and peak exercise (r=0.82; P <.001) were closely correlated.

Table 4. B-type Natriuretic Peptide (BNP) and N-terminal-pro-B-type Natriuretic Peptide (NT-proBNP) Kinetics
Reference Group (n=39)Pulmonary Limitation (n=28)Cardiocirculatory Limitation (n=27)P Value
BNP at rest (pg/mL)24(15-46)39(19-94)162(45-415)‡§<.001
BNP at peak exercise (pg/mL)33(15-56)⁎54(24-109)⁎219(32-484)⁎‡§<.001
ΔBNP (pg/mL)4(0-16)11(0-24)20(8-124)‡.007
% Δ BNP (%)10(0-76)22(0-43)18(2-53).81
NT-proBNP at rest (pg/mL)34(19-77)77(35-237)†506(129-1167)‡§<.001
NT-proBNP at peak exercise (pg/mL)34(27-85)⁎81(36-253)⁎†608(130-1204)⁎‡§<.001
ΔNT-proBNP (pg/mL)3(0-6)4(1-18)20(2-93)‡.006
% ΔNT-proBNP (%)8(0-17)5(2-8)7(1-10).29

ΔBNP/ΔNT-proBNP=absolute increase in BNP/NT-proBNP; %ΔBNP/%ΔNT-proBNP=relative increase in BNP/NT-proBNP.

Data are given as median (interquartile range).

Post hoc tests

⁎P <.05 compared with values at rest within the same group.

†P <.05 reference vs pulmonary limitation;

‡P <.05 reference vs cardiocirculatory limitation;

§P <.05 for pulmonary vs cardiocirculatory limitation.

Predictors of Cardiocirculatory Exercise Limitation 

In a first multivariate model with BNP as a covariate, arterial oxygen saturation (SaO2, P=.003), PaCO2 (P=.018), and BNP (P <.001) were independently associated with a cardiocirculatory limitation. In a second model with NT-proBNP as a covariate, a history of coronary artery disease (P=.045), SaO2 (P=.03), PaCO2 (P=.02), and NT-proBNP (P=.001) were independently associated with a cardiocirculatory limitation.

BNP and NT-proBNP at Rest for the Prediction of a Cardiocirculatory Limitation 

The area under the receiver operator characteristics curve (AUC) for BNP for the identification of a cardiocirculatory limitation was 0.79 (Figure 1). The optimal BNP cutoff of 85 pg/mL had a sensitivity (refers to the present selected population) of 63%, a specificity of 84%, a positive predictive value of 61%, and a negative predictive value of 85%. The corresponding AUC for NT-proBNP was 0.84 (Figure 1; z statistics for comparison with BNP 1.45; P=.15). The optimal NT-proBNP cutoff was 223 pg/mL and had a sensitivity of 74%, a specificity of 85%, a positive predictive value of 67%, and a negative predictive value of 89%.

  • View full-size image.
  • Figure 1. 

    Receiver operating characteristic (ROC) curves for the ability of B-type natriuretic peptide (BNP) and N-terminal-pro-B-type-natriuretic peptide (NT-proBNP) measured at rest to detect a cardiocirculatory limitation. The area under the ROC curve (AUC) with 95% confidence interval (CI) is given.

BNP and NT-proBNP at Peak Exercise for the Prediction of Cardiocirculatory Limitation 

The AUC for BNP at peak exercise to predict a cardiocirculatory limitation was 0.76 (Figure 2). The optimal BNP cutoff of 132 pg/mL had a sensitivity of 63% and a specificity of 88%. The AUC for NT-proBNP at peak exercise was 0.83 (Figure 2; z statistics for comparison with BNP 1.94; P=.052). The optimal NT-proBNP cutoff of 191 pg/mL had a sensitivity of 74% and a specificity of 84%. Peak BNP (z statistics −1.14; P=.25) and NT-proBNP (z statistics −0.75; P=.45) did not provide more diagnostic information than the corresponding levels at rest.

  • View full-size image.
  • Figure 2. 

    Receiver operating characteristic (ROC) curves for the ability of B-type natriuretic peptide (BNP) and N-terminal-pro-B-type-natriuretic peptide (NT-proBNP) measured at peak exercise to detect a cardiocirculatory limitation. The area under the ROC curve (AUC) with 95% confidence interval (CI) is given.

BNP and NT-proBNP Levels and Reduced Peak VO2 

BNP was related to the body weight indexed peak VO2 (r=−0.58; P <.001) as well as percent predicted peak VO2 (r=−0.35; P=.001), and NT-proBNP also was related to both body weight indexed peak VO2 (r=−0.66; P <.001) and percent predicted peak VO2 (r=−0.35; P=.001). Both BNP (AUC 0.67; 95% confidence interval 0.57-0.78; P=.003) and NT-proBNP (0.68; 95% confidence interval 0.58-0.79; P=.002), however, were only moderately accurate for the prediction of peak VO2 ≤84% of the predicted value, which is the cutoff for an impaired functional capacity according to guidelines.1

Relationship between BNP and NT-proBNP Key Cardiopulmonary Exercise Testing Variables 

Both BNP and NT-proBNP were negatively correlated to percent of age-predicted heart rate (BNP: r=−0.46, NT-proBNP: r=−0.45; P <.001 for both), ΔVO2/ΔWR (r=−0.34; P=.001, r=−0.31; P=.003), and oxygen pulse (r=−0.28; P=.008, r=−0.44; P <.001), and directly related to VD/VT at peak exercise (r=0.34; P=.001, r=0.45; P <.001) and peak VE/VCO2 (r=0.34; P <.001, r=0.50; P <.001). Breathing reserve was directly related to NT-proBNP (r=0.25; P=.02) and tended to be related to BNP (r=0.20; P=.06).

Value of BNP and NT-proBNP in Patients with Cardiac Disease 

There were 26 patients with known cardiac disease, of whom 5 had neither a cardiocirculatory nor a pulmonary limitation, 5 had pulmonary limitation, and 16 had cardiocirculatory limitation. Patients with (n=16) and without cardiocirculatory limitation (n=10) did not differ with respect to age (P=.32), body mass index (P=.59), and creatinine clearance (P=.57), but there was a significantly lower proportion of women in the group with cardiocirculatory limitation (5/10 vs 2/16; P=.036). There were trends towards higher BNP [259 (51-468) vs 157 (52-184) pg/mL; P=.097] and NT-proBNP [927 (261-1664) vs 288 (154-613) pg/mL; P=.12] levels in patients with cardiocirculatory limitation, but due to the small number of patients in this subgroup analysis, the differences did not reach statistical significance.

BNP and NT-proBNP in Patients Taking Beta-blockers 

Patients under beta-blocker therapy (n=25) were older, more likely to be male, to have diabetes, hypertension, cardiac disease, and pneumopathies other than chronic obstructive pulmonary disease and asthma, and to take cardiovascular medication; had lower left ventricular ejection fraction, percent predicted heart rate, percent predicted and body weight-indexed peak VO2, were more likely to have a cardiocirculatory limitation (P <.05 for all comparisons), and had higher BNP [162 (49-273) vs 26 (15-57); P <.001] and NT-proBNP [370 (168-956) vs 43 (23-99); P <.001] than patients without beta-blocker therapy.

The ability of BNP and NT-proBNP to predict a cardiocirculatory limitation was similar in patients taking or not taking beta-blockers. In those under beta-blocker therapy (n=25), the AUC for BNP and NT-proBNP were 0.77 (95% confidence interval [CI], 0.58-0.96) and 0.74 (0.54-0.94), respectively, and in those without beta-blocker therapy (n=69), the AUC for BNP and NT-proBNP were 0.69 (95% CI, 0.51-0.88) and 0.80 (95% CI, 0.64-0.95), respectively (P <.05 for all AUC).

Back to Article Outline

Discussion 

The present study evaluating the usefulness of BNP and NT-proBNP for the identification of a cardiocirculatory limitation in unselected patients with unexplained exercise intolerance revealed 3 new findings. First, patients with cardiocirculatory limitation had markedly higher resting BNP and NT-proBNP levels as compared with patients without. Second, the accuracy of BNP and NT-proBNP for the identification of a cardiocirculatory limitation was comparable. Third, additional measurements of BNP and NT-proBNP at peak exercise did not provide added information to measurements at rest.

This study extends and corroborates previous data on the use of BNP and NT-proBNP in settings with unmet needs regarding rapid and accurate diagnosis of common symptoms.2, 7, 8, 14 BNP and NT-proBNP seem to be markers of the burden of cardiac disease, thereby integrating the extent of left ventricular dysfunction, right ventricular dysfunction, and coronary disease.15 However, myocyte stretch seems not to be the only stimulus for BNP (and NT-proBNP, which is inactive, however) release,16 and thus BNP is not simply an imaging marker but a neurohormone17 reflecting the activation of the sympathetic nervous system and the renin-angiotension-aldosterone system. This is a particular strength of the 2 markers, as they may give an integrative view of cardiovascular function as cardiopulmonary exercise testing does. The AUC for BNP and NT-proBNP did not significantly differ, which may have been due to the comparatively small number of patients. Due to the longer half-life,6 NT-proBNP may have an advantage over BNP as a marker of a chronic condition such as a cardiocirculatory exercise limitation, because NT-proBNP may be less susceptible to short-term changes in cardiac filling pressure and other influences on BNP release and elimination.

It is evident that biomarkers will never be able to replace cardiopulmonary exercise testing in general. Similar to other settings,2, 5 however, the value of both biomarkers lies in their good negative predictive value, which might allow finding a clinical decision without performing cardiopulmonary exercise testing in certain cases. In contrast, if BNP and NT-proBNP are not low, cardiopulmonary exercise testing must be performed to exactly elaborate the mechanism underlying exercise intolerance. Of note, in approximately one third of participants, the type of exercise limitation could not be determined with sufficient certainty. One may speculate that BNP and NT-proBNP also might prove helpful if uncertainty persists after cardiopulmonary exercise testing.

With respect to the performance of the diagnostic test, it is important to note that the reference group did not consist of healthy people, but patients suffering from hyperventilation or patients with established cardiac or pulmonary disease in whom, based on cardiopulmonary exercise testing criteria, no pathological exercise limitation was identified. When looking at a subgroup of patients with known cardiac disease, BNP and NT-proBNP tended still to be higher in those with cardiocirculatory exercise limitation as compared with those without, although the difference did not reach statistical significance due to the small number of patients and the higher proportion of women in the group without cardiocirculatory limitation, and women are known to have higher BNP and NT-proBNP levels than men.18

The optimal cutoff values for the differentiation of presence vs. absence of a cardiocirculatory limitation for BNP (85 vs 100 pg/mL) and NT-proBNP (223 vs 300 pg/mL) were only slightly lower than those applied in the emergency department for the exclusion of acute heart failure.2, 5 Given that a difference in left ventricular filling pressures between the 2 settings is likely, one might expect a larger difference in the cutoffs. However, the correlation between BNP and left ventricular filling pressures at rest is generally weak.19, 20 Interestingly, NT-proBNP measured at rest has been shown to be strongly related to filling pressures during exercise but only weakly with values at rest in patients with suspected heart failure with preserved left ventricular ejection fraction.21

An increase in BNP and NT-proBNP during exercise has been in observed in healthy people and patients with cardiac disease,7, 8, 22, 23, 24, 25 although it is assumed that in contrast to A-type natriuretic peptide, the BNP precursor peptide is not stored in granules but regulated on the transcriptional levels. The mechanism underlying the exercise-induced BNP and NT-proBNP increase, as well as the diagnostic value, is unknown, however. In the present study, measurement of peak exercise BNP and NT-proBNP values did not provide added information to resting levels.

Our study has several limitations. First, we recruited consecutive patients referred to a university hospital. Thus, we cannot rule out some degree of referral bias. In addition, exclusion criteria created another bias in that excluded patients were older and thereby had lower exercise capacity. However, clearly defined criteria for a “good test” were required to ensure that cardiopulmonary exercise testing was really the reference standard to which natriuretic peptides were compared. As expected, the reference group was younger than the other 2 groups, and younger age per se is associated with lower BNP and NT-proBNP levels.18 However, such an age difference between the groups is typical in the clinical context, as the likelihood of significant cardiac disease increases with age.18 Second, echocardiographic data were not available for all patients. This also reflects the real-world setting of this study, however. Several patients were referred for an echocardiogram after cardiopulmonary exercise testing, as the test indicated a cardiocirculatory limitation. Third, patients with cardiocirculatory limitation had lower exercise capacity than patients with pulmonary limitation and controls, and both BNP and NT-proBNP were related to peak VO2. Thus, it remains to be shown whether BNP and NT-proBNP can identify cardiocirculatory limitation also in other settings.

Back to Article Outline

Conclusions 

Among patients referred for cardiopulmonary exercise testing for evaluation of unexplained exercise intolerance, BNP and NT-proBNP measured at rest were similarly useful to identify those with a cardiocirculatory limitation.

Back to Article Outline

Acknowledgment 

Diagnostic reagents were provided by Roche Diagnostics (Basel, Switzerland) and Abbott (Zug, Switzerland).

Back to Article Outline

References 

  1. American Thoracic Society; American College of Chest Physicians. ATS/ACCP Statement on cardiopulmonary exercise testing. Am J Respir Crit Care Med. 2003;167:211–277
  2. Maisel AS, Krishnaswamy P, Nowak RM, et al. Rapid measurement of B-type natriuretic peptide in the emergency diagnosis of heart failure. N Engl J Med. 2002;347:161–167
  3. Mueller C, Scholer A, Laule-Kilian K, et al. Use of B-type natriuretic peptide in the evaluation and management of acute dyspnea. N Engl J Med. 2004;350:647–654
  4. Januzzi JL, van Kimmenade R, Lainchbury J, et al. NT-proBNP testing for diagnosis and short-term prognosis in acute destabilized heart failure: an international pooled analysis of 1256 patients: the International Collaborative of NT-proBNP Study. Eur Heart J. 2006;27:330–337
  5. Januzzi JL, Camargo CA, Anwaruddin S, et al. The N-terminal Pro-BNP investigation of dyspnea in the emergency department (PRIDE) study. Am J Cardiol. 2005;95:948–954
  6. Daniels LB, Maisel AS. Natriuretic peptides. J Am Coll Cardiol. 2007;50:2357–2368
  7. Staub D, Jonas N, Zellweger MJ, et al. Use of N-terminal pro-B-type natriuretic peptide to detect myocardial ischemia. Am J Med. 2005;118:1287
  8. Staub D, Nusbaumer C, Zellweger MJ, et al. Use of B-type natriuretic peptide in the detection of myocardial ischemia. Am Heart J. 2006;151:1223–1230
  9. Schmid A, Schilter D, Fengels I, et al. Design and validation of an interpretative strategy for cardiopulmonary exercise tests. Respirology. 2007;12:916–923
  10. Guazzi M, Myers J, Arena R. Cardiopulmonary exercise testing in the clinical and prognostic assessment of diastolic heart failure. J Am Coll Cardiol. 2005;46:1883–1890
  11. Robbins M, Francis G, Pashkow FJ, et al. Ventilatory and heart rate responses to exercise : better predictors of heart failure mortality than peak oxygen consumption. Circulation. 1999;100:2411–2417
  12. Mueller T, Gegenhuber A, Poelz W, Haltmayer M. Preliminary evaluation of the AxSYM B-type natriuretic peptide (BNP) assay and comparison with the ADVIA Centaur BNP assay. Clin Chem. 2004;50:1104–1106
  13. Barnes SC, Collinson PO, Galasko G, et al. Evaluation of N-terminal pro-B type natriuretic peptide analysis on the Elecsys 1010 and 2010 analysers. Ann Clin Biochem. 2004;41:459–463
  14. Davis M, Espiner E, Richards G, et al. Plasma brain natriuretic peptide in assessment of acute dyspnoea. Lancet. 1994;343:440–444
  15. Mueller C, Breidthardt T, Laule-Kilian K, et al. The integration of BNP and NT-proBNP into clinical medicine. Swiss Med Wkly. 2007;137:4–12
  16. Martinez-Rumayor A, Richards AM, Burnett JC, Januzzi JL. Biology of the natriuretic peptides. Am J Cardiol. 2008;101:3–8
  17. Wermuth J, Staub D, Laule-Kilian K, et al. Neurohormonal activation and left ventricular ejection fraction in patients with suspected myocardial ischemia. Int J Cardiol. 2007;120:248–253
  18. Redfield MM, Rodeheffer RJ, Jacobsen SJ, et al. Plasma brain natriuretic peptide concentration: impact of age and gender. J Am Coll Cardiol. 2002;40:976–982
  19. O'Neill JO, Bott-Silverman CE, McRae AT, et al. B-type natriuretic peptide levels are not a surrogate marker for invasive hemodynamics during management of patients with severe heart failure. Am Heart J. 2005;149:363–369
  20. Forfia PR, Watkins SP, Rame JE, et al. Relationship between B-type natriuretic peptides and pulmonary capillary wedge pressure in the intensive care unit. J Am Coll Cardiol. 2005;45:1667–1671
  21. Tschope C, Kasner M, Westermann D, et al. Elevated NT-ProBNP levels in patients with increased left ventricular filling pressure during exercise despite preserved systolic function. J Card Fail. 2005;11:S28–S33
  22. Marie PY, Mertes PM, Hassan-Sebbag N, et al. Exercise release of cardiac natriuretic peptides is markedly enhanced when patients with coronary artery disease are treated medically by beta-blockers. J Am Coll Cardiol. 2004;43:353–359
  23. de Groote P, Soudan B, Lamblin N, et al. Is hormonal activation during exercise useful for risk stratification in patients with moderate congestive heart failure?. Am Heart J. 2004;148:349–355
  24. Barletta G, Stefani L, Del Bene R, et al. Effects of exercise on natriuretic peptides and cardiac function in man. Int J Cardiol. 1998;65:217–225
  25. Nicholson S, Richards M, Espiner E, et al. Atrial and brain natriuretic peptide response to exercise in patients with ischaemic heart disease. Clin Exp Pharmacol Physiol. 1993;20:535–540

 Funding: CM (Grant PP00B-102853) and MTM (Grant PBZHB-121007) are supported by the Swiss National Science Foundation.

 Conflicts of Interest: Dr. Mueller has received research support from Abbott, Biosite, Brahms, and Roche as well as speaker's honoraria from Abbott, Bayer, Biosite, Brahms, and Roche Diagnostic. The other authors have no possible conflicts of interest to declare. Diagnostic reagents were provided by Roche Diagnostics (Basel, Switzerland) and Abbott (Zug, Switzerland). The sponsors had no role in study design, data analysis, or interpretation.

 Authorship: All authors had access to the data and played a role in writing this manuscript.

PII: S0002-9343(08)00986-8

doi:10.1016/j.amjmed.2008.08.030

The American Journal of Medicine
Volume 122, Issue 3 , Pages 265-272, March 2009

Article Tools

* Image Usage & Resolution