Skip Navigation

European Journal of Heart Failure 2007 9(12):1178-1185; doi:10.1016/j.ejheart.2007.10.004
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Nielsen, O.W.
Right arrow Articles by McDonagh, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nielsen, O.W.
Right arrow Articles by McDonagh, T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2007 European Society of Cardiology

Value of BNP to estimate cardiac risk in patients on cardioactive treatment in primary care

O.W. Nielsenb,*, P.J. Cowburna, Ahmad Sajadiehc, J.J. Morton, H. Dargiea and T. McDonagha

a Cardiology Department, The Western Infirmary, Glasgow and MRC Clinical Research Initiative in Heart failure, Glasgow University United Kingdom
b Cardiology Department Y, Bispebjerg Hospital, University of Copenhagen 2400 Copenhagen NV, Denmark
c Cardiology Department, Amager Hospital, University of Copenhagen Denmark

* Corresponding author. Tel.: +45 35452141; fax: +45 3545 2568. own{at}dadlnet.dk (O.W. Nielsen).


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Funding
 References
 
Cardiac dysfunction may be suspected in patients with cardiovascular disease but identifying those with the highest risk is problematic. B-type natriuretic peptide (BNP) is a strong marker of heart failure in un-treated patients. This study evaluates a combined BNP and clinical algorithm for detecting cardiac dysfunction and the risk of death, in patients receiving cardioactive medication.

Methods: 459 stable general practice patients, who were taking typical heart failure drugs for any indication, were included. Echocardiography, ECG, and assay of NT-proANP and BNP (two methods) were performed. Regression models were used to identify items in a Risk Score to detect cardiac dysfunction.

Results: A Risk Score based on history of myocardial infarction (1 point), abnormal ECG (2 points), atrial fibrillation (1 point) and raised BNP (1–2 points) detected cardiac dysfunction with an AUC of ROC of 0.85. A Risk Score ≥2 had a sensitivity of 90%, specificity of 58%, and positive and negative predictive values of 37% and 96%. Risk Score and LVEF<0.36 also predicted mortality. Abnormal BNP defined as either > 100 pg/ml (Shionogi), or as age and sex related values, had similar predictive value.

Conclusion: In patients on cardioactive medication, a structured Risk Score based on raised BNP, history of MI, atrial fibrillation and abnormal ECG was useful for identifying patients for immediate further examination and those who could be evaluated later.

Key Words: Heart failure • Mortality • Diagnosis • Brain natriuretic • Cardiovascular treatment

Received March 4, 2007; Revised July 28, 2007; Accepted October 17, 2007


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 Funding
 References
 
Cardiac dysfunction may be suspected in any patient with cardiovascular disease [1] but the absolute risk is likely to be diverse. Many patients have a really low risk; however, there are patients who are at very high risk of having cardiac dysfunction or of dying. Patients with the highest risk would likely benefit from prompt cardiac examination, while others could be evaluated at some later occasion. The problem is how to identify those patients with the highest risk.

B-type brain natriuretic peptide (BNP) is a strong marker of heart failure and LV systolic dysfunction, but treatment with ACE-inhibitors and diuretics lowers BNP levels, which is likely to lower the prognostic and diagnostic value of BNP. In many cases, heart failure medication is already prescribed to patients at risk [1] of developing heart failure before appropriate investigations have been performed. It is therefore important to examine how BNP, in conjunction with other clinical data, can be used to detect cardiac dysfunction and estimate mortality in patients already on cardiovascular treatment.

Cardiac natriuretic peptide tests are useful in the diagnosis heart failure, but there is no clear consensus regarding the best cut-off value. A BNP cut-off value of 100 pg/ml (using the Biosite Triage assay) has been shown to significantly increase diagnostic accuracy in patients seen in the emergency department with acute shortness of breath [2,3]. However, at least 20-30% of out-patients with heart failure on optimal treatment have lower values [4]. In community populations lower cut-off values have been found to be relevant [5,6]. In the Framingham heart study values of 45 pg/ml (men) and 50 pg/ml (women) had a 95% specificity for LV systolic dysfunction, although a different BNP assay (Shionogi, Shionoria) was used in this study [7]. Some studies report that BNP levels should be evaluated according to the assay used, and the age and sex of the patients [8], and also that BNP levels may not be useful if patients are already on cardiovascular medication [9]. Thus, it is not clear how primary care physicians should use BNP testing for evaluating patients already on cardiovascular medication.

Most previous studies have examined the isolated diagnostic or prognostic value of BNP. Some other studies have combined BNP together with clinical data about ECG, history of myocardial infarction, diabetes, and atrial fibrillation in general practice [10-14] or the emergency room [15,16]. More evidence is now needed in patients taking cardioactive treatment, and it is relevant to consider both diagnosis and mortality as endpoints because a missed diagnosis of cardiac dysfunction may be acceptable if the patient is still alive to be evaluated later. In 1995-1996 we performed a population-based study of patients taking heart failure related drugs [17]. We now investigate a joint BNP and clinical strategy in the same population to identify the patients with the highest risk of having cardiac dysfunction or of dying. Our analysis is restricted to variables shown to be of importance in previous studies. The primary focus is to describe an algorithm that detects cardiac dysfunction and secondly to examine the risk of death in those patients that are "ruled in" or "ruled out" by the screening algorithm.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 Funding
 References
 
2.1. Patients
A heart failure nurse screened patient medication records at 13 randomly selected general practices (GP) in North Glasgow to identify all patients taking ACE-inhibitors, digoxin, or loop diuretic agents for any indication. The search identified 1194 patients. The nurse reviewed all GP case notes and asked the patient's own general practitioner if the patient would be able to attend for a cardiac examination in a hospital-based clinic. This consultation excluded 235 patients because they were either dead (n=25), too frail, hospitalised, had severe heart failure or were residing in a nursing home (n=173), had moved away from Glasgow (n=12) or had attended another screening study (n=25). The remaining 959 patients were randomly invited to attend for a cardiac examination until 500 patients had been studied [17]. All patients gave written informed consent to participate in the study.

2.2. Examinations
All subjects completed a self-reported questionnaire and underwent blood pressure measurement and standard 12-lead electrocardiograms (ECGs). Two cardiologists, who were blinded to all other patient data, independently coded the ECGs. An abnormal ECG was defined as clear evidence of a left bundle branch block (LBBB), wide QRS (>0.12 s), pathological Q-waves, major ST/T segment abnormality, or left ventricular hypertrophy (LVH). Borderline findings were adjudicated by consensus and patients with non-specific Q-waves or t-wave abnormalities were not included. Atrial fibrillation or sinus rhythm was noted separately.

2.3. Natriuretic peptides
Blood samples were taken after 20 min of supine rest. Samples were withdrawn into chilled tubes containing EDTA and trasylol (50 IU/ml) and centrifuged at 3000 g for 10 min at 4 °C, the separated plasma was stored at –20 °C prior to assay. For extraction of NT-proANP and BNP, the plasma was acidified with an equal volume of triflouroacetic acid, mixed, and centrifuged as previously described [6].

BNP was also measured directly in unextracted plasma using the Shionoria solid phase immunoradiometric assay from Schering CIS (France). In this assay, two monoclonal antibodies to BNP are used to ‘sandwich’ the BNP in plasma on a solid phase labelled with radioactive I125. The mean value quoted by the manufacturer for healthy men and women (n=420) is 5.9 pg/ml with 3SD upper limit of 18.4 pg/ml. The within assay coefficient of variation quoted is 2.0 and 2.7% for mean concentrations of 514 and 22.1 pg/ml respectively and between assay CV is 2.1 and 4.2% for concentrations of 525 and 21.1 pg/ml respectively. To convert from pg/ml to pmol/l divide by 3.467. The antibodies do not cross-react with any other natriuretic peptide fragments.

2.4. Cut-off values for BNP
Data are presented using the direct BNP method because these antibodies were similar to the Shionogi assay used in the Rochester Epidemiology Project and other studies. There is a general consensus that BNP values should be corrected for age and sex, we therefore used normal values from the Rochester study to categorise BNP into low normal (<75th percentile), high normal (from 75th to 95th percentile) and abnormal values (>95th percentile) in relation to sex and age [8]. The cut-off values in women <65 years were BNP <68 pg/ml (19.6 pmol/l) for low normal and BNP>192 pg/ml (55.4 pmol/l) for abnormal. The cut-off values in women >65 years were BNP<111 pg/ml (32.0 pmol/l) for low normal and BNP>233 pg/ml (67.2 pmol/l) for abnormal. The cut-off values in men <65 years were BNP<49 pg/ml (14.1 pmol/l) for low normal and BNP>146 pg/ml (42.1 pmol/l) for abnormal. The cut-off values in men >65 years of age were BNP<58 pg/ml (16.7 pmol/l) for low normal and BNP>177 pg/ml (51.1 pmol/l) for abnormal [8].

2.5. Echocardiography
Left ventricular ejection fraction (EF) was calculated by the Biplane Disc Summation Method (Simpson's Rule).[18] Left ventricular systolic function was also evaluated by the 9 segment model for assessing the wall motion index score (WMIs) [19,20], where WMIs of 1.2; 1.5; and 1.8 approximate to LVEF of 0.36, 0.45, and 0.54 respectively. Cardiac dysfunction in the multiple logistic regression analysis was defined as a LVEF<0.36 (by Simpsons biplane disc method or WMIs<1.2), or significant valve disease. Significant valve disease was decided by the cardiologist if the valve abnormality was judged to be of clinical relevance. Valve abnormality in association with atrial fibrillation was always considered as significant valve disease.

2.6. Statistics
2.6.1. Potential variables in the Risk Score
All natriuretic peptide values underwent logarithmic transformation before entering uni- and multivariate analyses. Univariate logistic regression models were used to calculate unadjusted odds ratios for each clinical variable to predict each of the endpoints: cardiac dysfunction, all cause mortality at three years and cardiovascular mortality during complete follow-up.

A multivariate logistic regression model identified the most important associates of cardiac dysfunction from previously recommended variables: sex (male=1, women=0), age/10 years, history of MI, abnormal ECG, history of diabetes, atrial fibrillation on ECG, (yes=1, no=0 for each variable), BNP in the three mentioned groups based on age and sex specific values (coded as 0-2). A p-value of <0.05 was required for inclusion. Covariates that were significantly associated with cardiac dysfunction in the multiple logistic regression model were used to create the Risk Score.

2.6.2. Final Risk Score for cardiac dysfunction
The contribution of each variable was proportional to the size of the beta coefficients from the multiple logistic regression model when cardiac dysfunction was used as the endpoint. A composite Risk Score was calculated for each patient and receiver-operating curves identified the score that gave a sensitivity of 90% for cardiac dysfunction.

2.6.3. Risk Score and mortality analyses
The prognostic value of the Risk Score was examined in Cox regression models. Flagging each patient's record with the Registrar General for Scotland on December 31, 2004 identified the date and cause of death. Assumptions of proportional hazards were assessed by visual judgment of the logarithm of minus logarithm of the survival estimates. Proportional hazard was seen for BNP both when divided into tertiles dichotomised around a value of 100 pg/ml and if divided into low, high or abnormal values as previously described.

First a Cox model examined sex, age, and the score value. Secondly, from this model we examined the additional value of the previously mentioned clinical and echocardiographic variables. Forward and backward selection of variables gave the same significant variables.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 Funding
 References
 
3.1. Patients and endpoints
A total of 459 patients had valid BNP data. The majority of patients were women who were on diuretic treatment (Table 1). Heart failure had been suspected by the GP in 147 patients (32%) of whom 88 were taking a loop diuretic. Cardiac dysfunction was observed in 21.4% of patients (n=98 of whom 53 had LVEF <0.36). Three-year mortality was 11.8% (n=54), and after 9 years of follow-up 36% of patients had died (n=166). 19.4% of patients died due to cardiovascular causes (n=89). Table 2 shows the association of each of the clinical variables to the different endpoints. For example, BNP was strongly associated with all types of endpoints and 59% of patients with abnormal BNP had cardiac dysfunction, compared to just 11% with a low normal BNP. Note cardiovascular deaths during complete follow-up were 45% in patients with abnormal BNP.


View this table:
[in this window]
[in a new window]

  		Table 1 Baseline characteristics of 459 patients with cardiovascular treatment

 


View this table:
[in this window]
[in a new window]

  		Table 2 Univariate associations of different clinical characteristics to three different endpoints

 
3.2. Univariate analyses
All predefined variables except diabetes were associated with cardiac dysfunction (Table 2), but hypertension was inversely associated with cardiac dysfunction and mortality. Atrial fibrillation was associated with cardiovascular deaths but not all cause mortality at 3 years.

3.3. Multivariate analyses
Indicators of cardiac dysfunction were atrial fibrillation on ECG (OR 2.3, 95%CI from 1.2 to 4.3, p=0.01), an abnormal ECG (OR 6.3, 95%CI from 3.2 to 12.5, p<0.001), a history of myocardial infarction (OR 2.9, 95%CI 1.7 to 4.9 p<0.001), and BNP in the three predefined groups (low normal, high normal and abnormal) (OR=2.2, 95%CI 1.3 to 3.2 per group, p<0.001). All peptides were associated with cardiac dysfunction. BNP values coded into the three predefined groups (low normal, high normal and abnormal) was significantly more related to cardiac dysfunction than using tertiles of BNP and more significant than using the extracted assay methods for NT-proANP and BNP. Surprisingly using BNP (direct method) values above or below 100 pg/ml (OR 3.8 with 95% CI from 2.2 to 6.6, p<0.001) was as significant as using BNP in the three groups (low, high and abnormal).

The model was not improved by other variables such as a history of angina, diabetes, hypertension, obesity, treatment with a beta-blocker, any diuretic, or loop diuretic. However, ACE-inhibitor use (p<0.001), and heart failure occurring during hospitalisation (p=0.066), were not surprisingly positively associated with cardiac dysfunction, while systolic blood pressure was inversely associated (p=0.04).

3.4. Derivation of Risk Score
The beta coefficients of the regression model were used to derive the scoring system (Table 3) to predict the risk of cardiac dysfunction. The composite score remained significant for predicting cardiac dysfunction (odds ratio 2.4 per point) and 3-year mortality (odds ratio 1.7 per point), when tested in a logistic regression model with other variables. This means that each of the original items did not add more information when the composite score was in the model.


View this table:
[in this window]
[in a new window]

  		Table 3 Risk Score to estimate risk of cardiac dysfunction defined as significant LV systolic dysfunction or significant valve disease

 
3.5. Performance of Risk Score
Fig. 1 illustrates that a score value of two gives about 10% probability of cardiac dysfunction and <10% probability of death while three points more than doubles the prevalence. The highest combined sensitivity and specificity from ROC analysis was obtained by a value of 3 or more yielding a sensitivity of 83% and specificity of 74%.


Figure 01
View larger version (15K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Probability of cardiac dysfunction and mortality at 3 years dependent on the Risk Score derived from Table 3.

 
Table 4 shows results of ROC analyses to diagnose cardiac dysfunction with a sensitivity of 90%. The Risk Score was superior to the other markers, and a composite score of 2 or more gave a sensitivity of 90% so we defined patients with 0-1 points as low risk, and points from 2 to 6 as high risk.


View this table:
[in this window]
[in a new window]

  		Table 4 Results of receiver-operating curve analyses of NT-proANP, BNP and Risk Score to detect cardiac dysfunction

 
Alternative ROC analyses examined 3-year mortality instead of cardiac dysfunction as the endpoint. The Risk Score had an AUC of 0.73 (95% CI from 0.66 to 0.79) while both BNP and NT-proANP gave AUC's of 0.66 (95%CI from 0.59 to 0.74). When cardiovascular mortality during complete follow-up was used as the endpoint then the AUC was 0.71 (95%CI from 0.66 to 0.77) for the Risk Score.

We also tested whether a single BNP cut-off value of 100 pg/ml would be better than using BNP in three groups. A logistic regression model then allocated one point to each of BNP>100 pg/ml, atrial fibrillation, history of myocardial infarction and 2 points for an abnormal ECG. ROC characteristics of this alternative score were very similar to the score from Table 2: AUC was 0.85, and a value of 2 or more gave a sensitivity of 91%, a specificity of 60%.

3.6. Multivariate survival modelling
The median follow-up time was 9.1 years (range 8.2 to 9.4 years). In a Cox regression model a Risk Score ≥2 was tested in association with age, sex, history of myocardial infarction, hypertension, diabetes, LVEF<0.36, valve disease, atrial fibrillation, BNP (low, high, abnormal), previous hospitalisation for heart failure. Then the independent significant variables were age per 10 years (Hazard ratio 1.60, p<0.001), LVEF<0.36 (HR 2.02, p=0.001), Risk Score≥2 (HR 1.67, p<0.006). There was no additional value of other variables although history of diabetes (HR 1.55, p=0.061) was of borderline significance. When short term mortality at three years was examined then LVEF<0.36 and Risk Score≥2 remained significant.

Table 5 shows the characteristics of the high and low risk patients. Among high risk patients nearly half had low normal BNP values, but 9-year mortality was 48.3% (115 patients), while cardiovascular mortality was 29.0% (69 patients). Almost half the patients had a low Risk Score and also a very low cardiovascular mortality of 9% (20 patients) after 9 years of follow-up. Just 4.5% of patients with a low Risk Score had cardiac dysfunction but cardiac dysfunction was not associated with short or long-term all cause mortality or cardiovascular mortality in patients with a low Risk Score.


View this table:
[in this window]
[in a new window]

  		Table 5 Distribution of characteristics in the low Risk Score (0-1 points) and high Risk Score (2-6 points) groups

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
Funding
 References
 
4.1. Principal findings
This study strongly suggests that it is more effective to use BNP in a structured Risk Score, than as a stand-alone test, to detect cardiac dysfunction in patients on cardioactive drug therapy. Further cardiac examination is indicated in patients with a Risk Score of 2 or more, while repeated risk evaluation every few years may be performed in patients with a low Risk Score. Furthermore, the presented Risk Score based on a crude BNP cut-off value of 100 pg/ml had the same performance as if age and sex related BNP values were used.

4.2. Performance of BNP in different high risk populations
There are many indications for echocardiography [21] and this Risk Score is a feasible method which could be used to help quantify clinical suspicion in this challenging group of patients. Our strategy performed better than in previous studies where BNP levels in patients on loop diuretics were unable to predict an LVEF<0.40 [9]. BNP performed better in studies of mixed normal and high risk subjects (mean AUC 0.83 of studies) [6,12,22-26], the general population (mean AUC 0.86) [7,12,23,24,27,28], and best of all in un-treated patients with new dyspnoea referred to rapid echocardiography or to the emergency department (mean AUC 0.93) [2,29,30], although the latter studies used heart failure as endpoint rather than a low LVEF.

4.3. BNP in relation to cardiovascular therapy
BNP is useful to aid diagnosis in patients admitted to hospital with acute dyspnoea. However, in stable out-patients BNP may be less accurate because of high use of both ACE-inhibitors and digoxin [17], poor renal function, atrial fibrillation [31], increased age and the presence of heart failure with preserved systolic function. In the ValHeFT study, the median BNP was 97 pg/ml (Shionogi assay) meaning that half the patients had lower values [32], but the prognostic value of baseline BNP was maintained independent of treatment [33].

4.4. BNP in relation to renal function and atrial fibrillation
It has been suggested that BNP decision cut-off values should be higher in patients with atrial fibrillation [31] and renal dysfunction[34]. Atrial fibrillation was an independent predictor of cardiac dysfunction in this study, partly because of its association with the definition of valve disease and also because evaluation of LV ejection fraction tends to generate lower values in cases of atrial fibrillation. Renal dysfunction elevates BNP concentrations and thus impacts on the diagnostic value of BNP. The importance of renal function was not examined in the present study, but renal dysfunction in some patients may enhance the need to use a Risk Score which takes into account risk factors other than BNP.

4.5. Diagnostic performance of the Risk Score
The likelihood ratios, predictive values and AUC's of the Risk Score are comparable to the best clinical variables in patients admitted with acute dyspnoea [35]. The prevalence of LV systolic dysfunction in treated patients is sufficiently high to merit screening or further investigation. It is therefore reassuring to know that a low score is associated with both a low prevalence of dysfunction as well as mortality. Furthermore, the Risk Score was significantly better than using BNP alone.

4.6. Application of the Risk Score in other populations
The Risk Score may slightly over-perform in the present cohort in which it was created; however, it should be relevant in other populations because it extends and improves on ideas developed in previous studies in primary care [10,11,13,22] and in the general population [6,7,12,24,36]. In contrast to some previous studies, the ECG was very useful possibly because we only counted major abnormalities whereas other studies also counted small abnormalities [26].

4.7. Limitations of the Risk Score when screening for cardiac dysfunction
The present Risk Score has been developed in stable and treated patients, whereas other score systems have been developed for acute heart failure. In acute heart failure scores were based on BNP and NT-proBNP with very different clinical items such as pulmonary oedema of chest X-ray, orthopnoea, absence of fever, loop diuretic use, age >75 years, rales, and absence of cough [15,16].

4.8. Strengths and weaknesses of the study
The present Risk Score findings are not likely to be a result of chance because the model was based on a few carefully selected variables which have been shown to be important in previous publications. The final variables were robustly identified independent of backward or forward selection and independent of order of entrance. However, performance of the Risk Score needs to be examined in other populations as well, and it is likely that the relative weight of the individual items may vary.

We used externally obtained reference values for BNP from a different population with a similar assay to avoid over fitting of data. It was somewhat surprising to find that a BNP value of 100 pg/ml performed at least as well in the Risk Score as age and sex adjusted BNP values.

Around the age of 70 years, a BNP of 48 pg/ml for men and 69 pg/ml for women denotes the 95th percentile [37]. The reason why sex and age related criteria may not be so important in this and other studies may be that significant LV systolic dysfunction induces values that are 300-500% higher than the median values in normal subjects and thus far more than the 40-60% difference induced by sex and age [23].

The Shionogi assay generates 50% higher values than the Biosite assays for values around 20 pg/ml [8], but at around 100 pg/ml the Biosite assay generates 20-30% higher values. The present Risk Score should therefore be tested with other assays in a similar context to develop comparable scoring systems. We did not examine NT-proBNP, but other studies have suggested cut-off values in both out-patient and hospital settings [25,38].

From a screening point of view, we also performed an analysis excluding patients previously hospitalised with heart failure. Results showed that our model tended to become stronger and therefore it is possible that the presented model may have underestimated the true value of the Risk Score in patients from primary care.

It is uncertain whether including heart failure with preserved systolic function as an endpoint would have improved the diagnostic value of our test [39]. BNP is slightly elevated in mild-moderate diastolic dysfunction but generally lower than in patients with systolic heart failure [40]. In a population study from Olmstead County, screening with NT-proBNP or BNP was not able to detect mild or moderate diastolic dysfunction [41].

4.9. Implications and future research
Our study shows that using a Risk Score including BNP for estimating risk of cardiac dysfunction and mortality is a rational method which may help primary care physicians to identify which patients should be referred immediately for further examination and those patients who could be re-evaluated in a few years. However, the outcome of such a strategy needs to be prospectively evaluated in clinical practice.


   Funding
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Funding
References
 
The study was funded by the British Heart Foundation and Medical Research Council, UK.


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Funding
 References
 

  1. Hunt H.A., Baker D.W., Chin M.H., et al. ACC/AHA Guidelines for the Evaluation and Management of Chronic Heart Failure in the Adult: Executive Summary A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1995 Guidelines for the Evaluation and Management of Heart Failure): Developed in Collaboration With the International Society for Heart and Lung Transplantation; Endorsed by the Heart Failure Society of America. Circulation (2001) 104(24):2996–3007.[Free Full Text]
  2. Maisel A., Krishnaswamy P., Nowak R.M., et al. Rapid measurement of B-type natriuretic peptide in the emergency diagnosis of heart failure. N Engl J Med (2002) 347(3):161–167.[Abstract/Free Full Text]
  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(7):647–654.[Abstract/Free Full Text]
  4. Tang W.H.W., Girod J.P., Lee M.J., et al. Plasma B-type natriuretic peptide levels in ambulatory patients with established chronic symptomatic systolic heart failure. Circulation (2003) 108(24):2964–2966.[Abstract/Free Full Text]
  5. Wang T.J., Larson M.G., Levy D., et al. Plasma natriuretic peptide levels and the risk of cardiovascular events and death. N Engl J Med (2004) 350(7):655–663.[Abstract/Free Full Text]
  6. McDonagh T.A., Robb S.D., Murdoch D.R., et al. Biochemical detection of left-ventricular systolic dysfunction. Lancet (1998) 351(9095):9–13.[CrossRef][Web of Science][Medline]
  7. Vasan R.S., Benjamin E.J., Larson M.G., et al. Plasma natriuretic peptides for community screening for left ventricular hypertrophy and systolic dysfunction: the Framingham heart study. JAMA (2002) 288(10):1252–1259.[Abstract/Free Full Text]
  8. Redfield M.M., Rodeheffer R.J., Jacobsen S.J., Mahoney D.W., Bailey K.R., Burnett J.C.J. Plasma brain natriuretic peptide concentration: impact of age and gender. J Am Coll Cardiol (2002) 40(5):976–982. JID- 8301365.[Abstract/Free Full Text]
  9. Hetmanski D.J., Sparrow N.J., Curtis S., Cowley A.J. Failure of plasma brain natriuretic peptide to identify left ventricular systolic dysfunction in the community. Heart (2000) 84(4):440–441.[Free Full Text]
  10. Nielsen O.W., Hilden J., Hansen J.F. Strong prognostic value of combining N-terminal atrial natriuretic peptide and ECG to predict death in heart patients from general practice. Heart (2001) 86(2):218–219.[Free Full Text]
  11. Nielsen O.W., Hansen J.F., Hilden J., Larsen C.T., Svanegaard J. Risk assessment of left ventricular systolic dysfunction in primary care: cross sectional study evaluating a range of diagnostic tests. BMJ (2000) 320(7229):220–224.[Abstract/Free Full Text]
  12. Nielsen O.W., McDonagh T.A., Robb S.D., Dargie H.J. Retrospective analysis of the cost-effectiveness of using plasma brain natriuretic peptide in screening for left ventricular systolic dysfunction in the general population. J Am Coll Cardiol (2003) 41(1):113–120.[Abstract/Free Full Text]
  13. Adlam D., Silcocks P., Sparrow N. Using BNP to develop a risk score for heart failure in primary care. Eur Heart J (2005) 26(11):1086–1093.[Abstract/Free Full Text]
  14. Cowie M.R., Jourdain P., Maisel A., et al. Clinical applications of B-type natriuretic peptide (BNP) testing. Eur Heart J (2003) 24(19):1710–1718.[Abstract/Free Full Text]
  15. Baggish A.L., Siebert U., Lainchbury J.G., et al. A validated clinical and biochemical score for the diagnosis of acute heart failure: the ProBNP Investigation of Dyspnea in the Emergency Department (PRIDE) Acute Heart Failure Score. Am Heart J (2006) 151(1):48–54.[CrossRef][Web of Science][Medline]
  16. McCullough P.A., Nowak R.M., McCord J., et al. B-type natriuretic peptide and clinical judgment in emergency diagnosis of heart failure: analysis from Breathing Not Properly (BNP) Multinational Study. Circulation (2002) 106(4):416–422.[Abstract/Free Full Text]
  17. Nielsen O.W., McDonagh T., Cowburn P., Blue L., Robb S.D., Dargie H. Patient differences related to management in general practice and the hospital: a cross-sectional study of heart failure in the community. Eur Heart J (2004) 25(19):1718–1725.[Abstract/Free Full Text]
  18. Schiller N.B., Acquatella H., Ports T.A., et al. Left ventricular volume from paired biplane two-dimensional echocardiography. Circulation (1979) 60(3):547–555.[Abstract/Free Full Text]
  19. Heger J.J., Weyman A.E., Wann L.S., Rogers E.W., Dillon J.C., Feigenbaum H. Cross-sectional echocardiographic analysis of the extent of left ventricular asynergy in acute myocardial infarction. Circulation (1980) 61(6):1113–1118.[Free Full Text]
  20. Berning J., Rokkedal N.J., Launbjerg J., Fogh J., Mickley H., Andersen P.E. Rapid estimation of left ventricular ejection fraction in acute myocardial infarction by echocardiographic wall motion analysis. Cardiology (1992) 80(3-4):257–266.[Web of Science][Medline]
  21. Khunti K. Systematic review of open access echocardiography for primary care. Eur J Heart Fail (2004) 6(1):79–83.[Abstract/Free Full Text]
  22. Fuat A., Murphy J.J., Hungin A.P., et al. The diagnostic accuracy and utility of a B-type natriuretic peptide test in a community population of patients with suspected heart failure. Br J Gen Pract (2006) 56(526):327–333.[Web of Science][Medline]
  23. Galasko G.I., Lahiri A., Barnes S.C., Collinson P., Senior R. What is the normal range for N-terminal pro-brain natriuretic peptide? How well does this normal range screen for cardiovascular disease? Eur Heart J (2005) 26(21):2269–2276.[Abstract/Free Full Text]
  24. Hobbs F.D., Davis R.C., Roalfe A.K., Hare R., Davies M.K., Kenkre J.E. Reliability of N-terminal pro-brain natriuretic peptide assay in diagnosis of heart failure: cohort study in representative and high risk community populations. BMJ (2002) 324(7352):1498.[Abstract/Free Full Text]
  25. Gustafsson F., Steensgaard-Hansen F., Badskjaer J., Poulsen A.H., Corell P., Hildebrandt P. Diagnostic and prognostic performance of N-terminal ProBNP in primary care patients with suspected heart failure. J Card Fail (2005) 11(5 Suppl):S15–S20.[CrossRef][Web of Science][Medline]
  26. Zaphiriou A., Robb S., Murray-Thomas T., et al. The diagnostic accuracy of plasma BNP and NTproBNP in patients referred from primary care with suspected heart failure: results of the UK natriuretic peptide study. Eur J Heart Fail (2005) 7(4):537–541.[Abstract/Free Full Text]
  27. Smith H., Pickering R.M., Struthers A., Simpson I., Mant D. Biochemical diagnosis of ventricular dysfunction in elderly patients in general practice: observational study. BMJ (2000) 320(7239):906–908.[Abstract/Free Full Text]
  28. Groenning B.A., Raymond I., Hildebrandt P.R., Nilsson J.C., Baumann M., Pedersen F. Diagnostic and prognostic evaluation of left ventricular systolic heart failure by plasma N-terminal pro-brain natriuretic peptide concentrations in a large sample of the general population. Heart (2004) 90(3):297–303.[Abstract/Free Full Text]
  29. Nielsen L.S., Svanegaard J., Klitgaard N.A., Egeblad H. N-terminal pro-brain natriuretic peptide for discriminating between cardiac and non-cardiac dyspnoea. Eur J Heart Fail (2004) 6(1):63–70.[Abstract/Free Full Text]
  30. Cowie M.R., Struthers A.D., Wood D.A., et al. Value of natriuretic peptides in assessment of patients with possible new heart failure in primary care. Lancet (1997) 350(9088):1349–1353.[CrossRef][Web of Science][Medline]
  31. Knudsen C.W., Omland T., Clopton P., et al. Impact of atrial fibrillation on the diagnostic performance of B-type natriuretic peptide concentration in dyspneic patients: an analysis from the breathing not properly multinational study. J Am Coll Cardiol (2005) 46(5):838–844.[Abstract/Free Full Text]
  32. Latini R., Masson S., Anand I., et al. Effects of valsartan on circulating brain natriuretic peptide and norepinephrine in symptomatic chronic heart failure: the Valsartan Heart Failure Trial (Val-HeFT). Circulation (2002) 106(19):2454–2458.[Abstract/Free Full Text]
  33. Latini R., Masson S., Anand I., et al. The comparative prognostic value of plasma neurohormones at baseline in patients with heart failure enrolled in Val-HeFT. Eur Heart J (2004) 25(4):292–299.[Abstract/Free Full Text]
  34. McCullough P.A., Duc P., Omland T., et al. B-type natriuretic peptide and renal function in the diagnosis of heart failure: an analysis from the Breathing Not Properly Multinational Study. Am J Kidney Dis (2003) 41(3):571–579.[CrossRef][Web of Science][Medline]
  35. Wang C.S., FitzGerald J.M., Schulzer M., Mak E., Ayas N.T. Does this dyspneic patient in the emergency department have congestive heart failure? JAMA (2005) 294(15):1944–1956. %19.[Abstract/Free Full Text]
  36. Ng L.L., Loke I., Davies J.E., et al. Identification of previously undiagnosed left ventricular systolic dysfunction: community screening using natriuretic peptides and electrocardiography. Eur J Heart Fail (2003) 5(6):775–782.[Abstract/Free Full Text]
  37. Wang T.J., Larson M.G., Levy D., et al. Impact of age and sex on plasma natriuretic peptide levels in healthy adults. Am J Cardiol (2002) 90(3):254–258.[CrossRef][Web of Science][Medline]
  38. Januzzi J.L. Jr., Camargo C.A., Anwaruddin S., et al. The N-terminal Pro-BNP investigation of dyspnea in the emergency department (PRIDE) study. Am J Cardiol (2005) 95(8):948–954.[CrossRef][Web of Science][Medline]
  39. Lubien E., DeMaria A., Krishnaswamy P., et al. Utility of B-natriuretic peptide in detecting diastolic dysfunction: comparison with Doppler velocity recordings. Circulation (2002) 105(5):595–601.[Abstract/Free Full Text]
  40. O'donoghue M., Chen A., Baggish A.L., et al. The effects of ejection fraction on N-terminal ProBNP and BNP levels in patients with acute CHF: analysis from the ProBNP Investigation of Dyspnea in the Emergency Department (PRIDE) study. J Card Fail (2005) 11(5 Suppl):S9–S14.[CrossRef][Web of Science][Medline]
  41. Costello-Boerrigter L.C., Boerrigter G., Redfield M.M., et al. Amino-terminal pro-B-type natriuretic peptide and B-type natriuretic peptide in the general community: determinants and detection of left ventricular dysfunction. J Am Coll Cardiol (2006) 47(2):345–353.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Eur J Heart FailHome page
J. Pimenta, P. Bettencourt, and A. Azevedo
Validation of a risk score to estimate cardiac risk in subjects from the general population on cardioactive treatment
Eur J Heart Fail, June 1, 2008; 10(6): 621 - 622.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (3)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Nielsen, O.W.
Right arrow Articles by McDonagh, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nielsen, O.W.
Right arrow Articles by McDonagh, T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?