Post on 29-Apr-2023
doi: 10.1111/j.1365-2796.2009.02135.x
Comparison of midregional pro-atrial natriureticpeptide with N-terminal pro-B-type natriureticpeptide in the diagnosis of heart failure
M. Potocki1,5, T. Breidthardt1, T. Reichlin1, S. Hartwiger1, N. G. Morgenthaler2, A. Bergmann2, M. Noveanu1,H. Freidank3, A. B. Taegtmeyer1, K. Wetzel1, T. Boldanova1, C. Stelzig1, R. Bingisser1, M. Christ4 & C. Mueller1
From the 1Department of Internal Medicine, University Hospital, Basel, Switzerland; 2Research Department, B.R.A.H.M.S. AG,Hennigsdorf ⁄Berlin, Germany; 3Department of Laboratory Medicine, University Hospital, Basel, Switzerland; 4KlinikumNuernberg, Internal Medicine, Nuernberg; and 5Herz-Zentrum Bad Krozingen, Bad Krozingen, Germany
Abstract. Potocki M, Breidthardt T, Reichlin T,
Hartwiger S, Morgenthaler NG, Bergmann A,
Noveanu M, Freidank H, Taegtmeyer AB, Wetzel K,
Boldanova T, Stelzig C, Bingisser R, Christ M,
Mueller C (University Hospital, Basel, Switzerland,
B.R.A.H.M.S. AG, Hennigsdorf ⁄Berlin, Germany;
Klinikum Nuernberg, Internal Medicine, Nuernberg;
and Herz-Zentrum Bad Krozingen, Bad Krozingen,
Germany). Comparison of midregional pro-atrial natri-
uretic peptide with N-terminal pro-B-type natriuretic
peptide in the diagnosis of heart failure. J Intern Med
2010; 267: 119–129.
Objectives. The concentration of atrial natriuretic pep-
tide (ANP) in the circulation is approximately 10- to
50- fold higher than B-type natriuretic peptide (BNP).
We sought to compare the accuracy of midregional
pro-atrial natriuretic peptide (MRproANP) measured
with a novel sandwich immunoassay with N-terminal
pro-B-type natriuretic peptide (NTproBNP) in the
diagnosis of heart failure.
Design. The diagnosis of heart failure was adjudicated
by two independent cardiologists using all available
clinical data (including BNP levels) in 287 consecutive
patients presenting with dyspnoea to the emergency
department (ED). MRproANP and NTproBNP levels
were determined at presentation in a blinded fashion.
Results. Heart failure was the adjudicated final diagno-
sis in 154 patients (54%). Median MRproANP was
significantly higher in patients with heart failure as
compared to patients with other causes of dyspnoea
(400 vs. 92 pmol L)1, P < 0.001). The diagnostic
accuracy of MRproANP was very high with an area
under the receiver operating characteristic curve of
0.92 and was comparable with that of NTproBNP
(0.92, P = 0.791). Moreover, MRproANP provided
incremental diagnostic information to BNP and
NTproBNP in patients presenting with BNP levels in
the grey zone between 100 and 500 pg mL)1.
Conclusion. Midregional pro-atrial natriuretic peptide is
as accurate in the diagnosis of heart failure as
NTproBNP. MRproANP seems to provide incremental
information on top of BNP or NT-proBNP in some
subgroups and should be further investigated in other
studies.
Keywords: dyspnoea, emergency diagnosis, heart fail-
ure, MRproANP, natriuretic peptides.
Introduction
Heart failure (HF) is common, associated with high
morbidity and mortality, and difficult to diagnose,
particularly, in the emergency department (ED) [1, 2].
Dyspnoea is the leading symptom of most HF
patients. Unfortunately, neither patient history nor
physical examination can accurately differentiate
ª 2009 Blackwell Publishing Ltd 119
Original Article |
dyspnoea because of HF from dyspnoea due to other
causes, such as pulmonary diseases [3, 4]. However,
accurate diagnosis is necessary for the selection of the
most appropriate treatment.
B-type natriuretic peptides are quantitative markers of
HF that have been shown to be very helpful in the
diagnosis of HF. The use of B-type natriuretic peptide
(BNP) and its amino-terminal fragment, N-terminal
pro-B-type natriuretic peptide (NTproBNP), signifi-
cantly increases the diagnostic accuracy in the ED [5,
6] and thereby improves patient evaluation and treat-
ment [7–9].
The concentration of atrial natriuretic peptide (ANP)
in the circulation is approximately 10- to 50- fold
higher than that of BNP [10]. Therefore, the biologi-
cal signal reflected by the increased ANP may be
pathophysiologically and therefore diagnostically even
more important than the signal of BNP. However,
analytical difficulties with previous ANP assays and
inferior clinical results as compared with BNP in
early pilot studies have hampered and distracted fur-
ther research on the diagnostic use of ANP and its
precursors [11]. Mature ANP is derived from the pre-
cursor N-terminal-proANP (NTproANP), which is sig-
nificantly more stable in the circulation than the
mature peptide and is therefore thought to be a more
reliable substrate for analysis [12]. Furthermore, since
NTproANP has been shown to be subject of further
fragmentation [13, 14], immunoassays for measure-
ment of a mid-regional sequence of proANP
(MRproANP) may have an advantage [15]. We there-
fore determined plasma levels of MRproANP in unse-
lected patients presenting to the ED with dyspnoea
and compared the accuracy of MRproANP in the
diagnosis of HF with that of NTproBNP.
Methods
Study population and design
We prospectively enrolled 287 consecutive patients pre-
senting to the ED of the University Hospital Basel with
dyspnoea as the most prominent symptom from April
2006 to March 2007. Patients under 18 years of age,
patients on haemodialysis and trauma patients were
excluded. The study was carried out according to the
principles of the Declaration of Helsinki and approved
by the local ethics committee. Written informed consent
was obtained from all participating patients.
Clinical evaluation of patients
Patients underwent an initial clinical assessment that
included clinical history, physical examination, ECG,
pulse oximetry, blood tests and chest X-ray. For each
patient enrolled in the study, ED physicians assessed
the probability that the patient had HF (by assigning a
value of 0% to 100% clinical certainty) as the cause
of dyspnoea using all available information from the
physical examination, ECG, chest X-ray and blood
tests including BNP. BNP values were available at
presentation in all enrolled patients.
Reference (‘Gold’) standard definition of heart failure
Two independent cardiologists reviewed all medical
records pertaining to the patient and classified the diag-
nosis as dyspnoea caused by HF or dyspnoea because
of causes other than HF. Both cardiologists had access
to all available medical records pertaining to the patient
from the time of ED presentation to the results of the
90-day follow up. This information included BNP lev-
els, chest X-ray, medical history not available at the
time of admission, echocardiography, left ventriculog-
raphy (performed at the time of cardiac catheterization),
pulmonary function test, CT scan, right heart catheteri-
zation, hospital course, response to therapy, autopsy
data for deceased patients and information about clini-
cal events or readmissions during the 90-day follow up
period. The BNP level was considered as a quantitative
marker of HF and therefore interpreted as a continuous
variable to make best use of the information provided
by this test. Absolute BNP values were adjusted for the
presence of both kidney disease and obesity to maxi-
mize diagnostic accuracy [16].
Measurement of natriuretic peptides
At presentation to the ED, blood samples for determi-
nation of MRproANP and NTproBNP levels were col-
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
120 ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129
lected into tubes containing potassium EDTA. After
centrifugation, samples were immediately frozen at
)80 �C until later analysis. Detection of MRproANP
was performed using a sandwich immunoassay
(MRproANP LIA, B.R.A.H.M.S, Hennigsdorf ⁄Berlin,Germany) as described in detail elsewhere [15]. The
functional assay sensitivity (interassay coefficient of
variance <20%) is 20 pmol L)1. The stability of
MRproANP at room temperature is >24 h. After 7 days
at room temperature, the degradation of MRproANP
does not exceed 20%. This assay allows measurement
of MRproANP in serum and plasma (with EDTA, hepa-
rin, or citrate) [14]. Median MRproANP in 325 healthy
individuals in previous investigations was 45 pmol L)1
(95% confidence interval (CI) 43 to 49 pmol L)1) [15].
NTproBNP levels were determined by a quantitative
electrochemiluminescence immunoassay (Elecsys pro-
BNP; Roche Diagnostics AG, Zug, Switzerland). BNP
was measured by a microparticle enzyme immunoassay
(AxSym; Abbott Laboratories, Abbott Park, IL, USA).
End-points
The primary end-point was to compare the accuracy
of MRproANP with that of NTproBNP to diagnose
HF. As BNP levels were available for the reference
(‘Gold’) standard diagnosis of HF, the comparison of
the MRproANP with BNP was considered biased in
favour of BNP and, therefore, the comparisons were
made only between MRproANP and NTproBNP. The
secondary end-point was to investigate, whether
the addition of MRproANP to clinical judgement in the
ED, including chest X-ray and BNP information,
would further increase diagnostic accuracy.
Statistical analysis
Univariate data on demographic and clinical features
were compared by Mann–Whitney U test or Fisher’s
exact test as appropriate. Medians of each marker in
those with and without HF were compared using non-
parametric testing; otherwise, MRproANP and
NTproBNP were log-transformed to achieve a normal
distribution. Receivers operating characteristic (ROC)
curves were utilized to evaluate the utility of
MRproANP and NTproBNP for the diagnosis of HF
and areas under the curve (AUCs) were calculated for
both markers. AUCs were compared according to the
method by Hanley and McNeil [17]. The optimal cut-
off point was determined by selecting the point on the
ROC curve that maximized both sensitivity and speci-
ficity. The optimum cut-off point for ED clinical cer-
tainty of HF was chosen at ‡80%, a cut-off point
providing reasonable certainty and practical applicabil-
ity [18]. Logistic regression was used to combine clini-
cal judgement with MRproANP data in predicting final
diagnosis. The diagnostic accuracies of both assays at
optimal cut-off concentrations as determined by ROC
analysis were compared by the McNemar test (compari-
son of discordant pairs of false biochemical classifica-
tions). To determine odds ratios for the detection of HF
with respect to MRproANP and NTproBNP cut-off
points of highest diagnostic accuracy, logistic regres-
sion analysis was performed unadjusted and adjusted
for significant covariates in univariate analysis (age,
prior history of HF, prior myocardial infarction, esti-
mated glomerular filtration rate, haemoglobin, pulmo-
nary rales, peripheral oedema and jugular venous
distension). Spearman’s coefficient of rank correlation
was used to assess the relation of MRproANP and
NTproBNP concentrations in the study population.
Body mass index (BMI) was calculated using the con-
ventional formula of weight in kilograms divided by
the square of height in meters. Normal weight was
defined as a BMI <25 kg m)2, overweight as a BMI
between 25 kg m)2 and 29.9 kg m)2 and Obesity as a
BMI of 30 kg m)2 or greater, according to the defini-
tion of the World Health Organization. Glomerular fil-
tration rate was calculated using the Modification of
Diet in Renal Disease (MDRD) formula [19]. Data
were statistically analysed with SPSS 15.0 software
(SPSS Inc, Chicago, IL, USA) and the MedCalc 9.3.9.0
package (MedCalc Software, Mariakerke, Belgium).
All probabilities were two-tailed and P < 0.05 was con-
sidered to indicate statistical significance.
Results
Study population
A total of 287 patients were enrolled in this study
(Table 1). The median age was 77 years and there
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129 121
Table 1 Patients characteristics
Characteristic All patients (n = 287)
Heart failure
(n = 154, 54%)
No heart failure
(n = 133, 46%) P-value
Age (years)b 77 (68–83) 80 (75–85) 71 (58–79) <0.001
Male gender (%) 52 51 53 0.906
Body mass index (kg m)2)a 26 ± 6 27 ± 6 26 ± 7 0.124
History (%)
Hypertension 68 78 56 <0.001
History of heart failure 24 40 7 <0.001
Coronary artery disease 28 38 16 <0.001
Diabetes mellitus 18 24 11 0.006
Chronic obstructive pulmonary disease 34 27 42 0.006
Chronic kidney disease 28 44 11 <0.001
Shortness of breath (%)
Whilst walking up a slight incline 20 9.7 32 <0.001
Whilst walking on level ground 40 45 35 0.109
At rest 40 46 33 0.034
Physical examination findings (%)
Heart rate (bpm)a 93 ± 23 93 ± 25 92 ± 19 0.495
Systolic blood pressure (mmHg)a 138 ± 26 135 ± 27 140 ± 25 0.098
Diastolic blood pressure (mmHg)a 83 ± 16 82 ± 17 83 ± 14 0.455
Respiratory rateb 24 (20–28) 24 (20–28) 24 (18–28) 0.678
Rales 54 64 43 0.001
Lower extremity oedema 42 57 26 <0.001
Hepatojugular reflux 8.0 12 3.8 0.016
Jugular venous distension 28 44 11 <0.001
Chronic Medication (%)
b-blockers 39 57 17 <0.001
Angiotensin-converting enzyme
inhibitor ⁄ angiotensin-receptor blockers49 62 34 <0.001
Loop diuretics 52 64 38 <0.001
Calcium antagonists 18 21 14 0.070
Digoxin 4.9 5.2 4.5 0.169
Spironolactone 2.4 3.9 0.8 0.041
Laboratory findings
Serum creatinine (lmol L)1)b 85 (66–120) 99 (79–147) 71 (56–96) <0.001
eGFR (mL min)1 1.73m)2)b 67 (44–89) 54 (36–73) 80 (63–112) <0.001
Blood urea nitrogen (mmol L)b 7.3 (5.4–12.0) 9.7 (6.7–15.6) 6.0 (4.4–7.7) <0.001
Sodium (mmol L)b 137 (134–139) 137 (135–139) 137 (134–140) 0.238
Haemoglobin (g L)b 133 (118–145) 129 (112–141) 138 (125–150) <0.001
Troponin T (lg L)b (n = 192) 0.01 (0.01–0.03) 0.01 (0.01–0.04) 0.01 (0.01–0.01) <0.001
BNP (pg mL)1)b 349 (90–1120) 976 (467–1925) 81 (39–181) <0.001
NT-proBNP (pg mL)1)b 1656 (314–6105) 5757 (1924–13 243) 300 (76–974) <0.001
MR-proANP (pmol L)1)b 221 (93–441) 400 (246–643) 92 (60–173) <0.001
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
122 ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129
were 149 men (52%). Arterial hypertension was
present in 68% of the patients, 28% had coronary
heart disease and 34% had chronic obstructive pul-
monary disease. On examination, 54% of patients
had pulmonary rales, 42% had lower extremity
oedema and 28% had jugular venous distension.
Diuretics (52%) were the most common prescribed
chronic medications, followed by ACE-inhibitors or
angiotensin-receptor-blockers (49%) and b-blockers(39%).
Final (gold standard) diagnosis
The adjudicated gold standard diagnosis was HF in
154 (54%) patients and no HF in 133 patients.
Patients without HF had chronic obstructive pulmo-
nary disease (57 patients), pneumonia (32 patients),
pulmonary embolism (8 patients), malignancy (7
patients), hyperventilation (5 patients) and other
causes of dyspnoea such as interstitial lung disease,
asthma or bronchitis (24 patients).
Natriuretic peptide levels and diagnosis of HF
The levels of MRproANP and NTproBNP in patients
with and without an adjudicated gold standard diagno-
sis of HF are shown in Fig. 1. The median
MRproANP concentration of patients with HF
(400 pmol L)1, IQR 246–642 pmol L)1) was signifi-
cantly higher than of those without HF (92 pmol L)1,
IQR 92–173 pmol L)1; P < 0.001 for difference). The
median NTproBNP concentration in patients with HF
was 5757 pg mL)1 (IQR 1924–13243 pg mL)1) vs.
300 pg mL)1 (IQR 76–974 pg mL)1) in patients
without HF (P < 0.001 for difference). Spearman’s
coefficient of rank correlation was 0.899 between
MRproANP and NTproBNP (P < 0.001).
The AUC of MRproANP was 0.92 (95% CI 0.88 to
0.95) for the diagnosis of HF, comparable to the AUC
of NTproBNP with 0.92 (95% CI 0.89 to 0.95;
P = 0.791; Fig. 2a). On the basis of the ROC analy-
sis, the optimal diagnostic cut-off value for
MRproANP and NTproBNP to reach the highest diag-
nostic accuracy was 206 pmol L)1 and 1540 pg mL)1
respectively. At this cut-off, diagnostic accuracy, sen-
sitivity and specificity for MRproANP were all 84%.
Logistic regression analysis confirmed MRproANP as
a powerful predictor of HF in both univariate and
multivariate analyses (Table 2).
The AUC for the combination of ED-probability with
MRproANP was significantly higher than for ED-
probability alone (P = 0.016, Fig. 2b). The diagnostic
accuracy for clinical judgement with high ED-probabil-
Table 1 Continued
Characteristic All patients (n = 287)
Heart failure
(n = 154, 54%)
No heart failure
(n = 133, 46%) P-value
Echocardiography findings (%) (n = 116) (n = 89) (n = 27)
Left ventricular ejection fractionb 56 (35–65) 50 (33–60) 65 (60–65) <0.001
eGRF, estimated glomerular filtration rate; MRproANP, midregional pro-atrial natriuretic peptide; NTproBNP, N-terminal pro-B-type natriureticpeptide.aMeans ± SD. bMedian (25th–75th percentile).
NT
-pro
BN
P c
once
ntra
tion
(pg/
mL)
35 000
30 000
25 000
20 000
15 000
10 000
5000
0
MR
-proAN
P concentration (pm
ol/L)
1500
1200
900
600
300
0
MR-proANP NT-proBNP
Acute HF No acute HF Acute HF No acute HF
Fig. 1 The MRproANP and NTproBNP levels in patientswith and without heart failure – NTproBNP and MRproANPlevels for patients with dyspnoea caused by acute heart failure(HF) (n = 154) and patients with dyspnoea attributable toother causes (n = 133); all P < 0.001. Boxes represent the25th and 75th percentile, whereas whiskers represent theminimum to the maximum concentration, excluding outliers,which are displayed as separate points.
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129 123
ity of HF (80% to 100% clinical certainty) was 79.4%.
For a composite decision based on ED-probability of
80% to 100% or MRproANP >206 pmol L)1, or both,
the diagnostic accuracy was 87.8%. Therefore,
MRproANP levels >206 pmol L)1 added to clinical
judgement would have increased the diagnostic accu-
racy considerably to almost 88% (P < 0.001) and
increased the sensitivity from 62% to 91%. A similar
improvement was observed for the combination of clin-
ical judgement and NTproBNP, resulting in a diagnos-
tic accuracy of 86.4%. Looking at the patients who
were additionally identified as having HF by
MRproANP compared to the other patients with heart
failure, we identify differences in following clinical
characteristics: These patients more often had an under-
lying chronic obstructive pulmonary disease (38% vs.
22%, P = 0.044), had less clinical signs of congestion
like rales (51% vs. 70%, P = 0.028), showed a better
ejection fraction (60[50–65]% vs. 41[28–60]%,
P = 0.012) and were more often on medical therapy
with ACE-Inhibitors or angiotensin-receptor-blockers
(76% vs. 57%, P = 0.03). Table 3 gives detailed diag-
nostic information including the appropriate decision
statistics for the clinical and biochemical diagnosis
of HF.
Subgroup analysis
The additional value of MRproANP in several impor-
tant subgroups where BNP and NTproBNP levels are
difficult to interpret is shown in Table 4. We focused
on patients with BNP values between 100 and
500 pg mL)1, GFR values <60 mL min)1 and patient
with BMI >30 kg m)2. In each of these subgroups of
patients, MR-proANP was added to a logistic regres-
sion model with BNP or NTproBNP to predict HF. In
patients with BNP values between 100 and
500 pg mL)1, MRproANP added significant addi-
tional information to BNP and NTproBNP (OR 8.9,
P = 0.022; respectively OR 9.8, P = 0.042).
NTproBNP showed no additional information to BNP
(OR 1.6, P = 0.37). For patients with BMI
>30 kg m)2, MRproANP showed only a trend for an
incremental effect on top of BNP or NTproBNP
(P = 0.063 and P = 0.084 respectively).
The associations of MRproANP, NTproBNP and BMI
were examined in patients with and without heart
failure. Especially in patients with heart failure,
1 - specificity1.00.80.60.40.20.0
sen
siti
vity
1.0
0.8
0.6
0.4
0.2
0.0
P = 0.791
NT-proBNP (AUC = 0.92)MR-proANP (AUC = 0.92)
1 - specificity1.00.80.60.40.20.0
sen
siti
vity
1.0
0.8
0.6
0.4
0.2
0.0
P = 0.016
ED-Probability combined with MR-proANP (AUC = 0.96)
MR-proANP (AUC = 0.92)ED-Probability (AUC = 0.90)
(a)
(b)
Fig. 2 (a) Receiver operating characteristic curves forNTproBNP and MRproANP to diagnose heart failure inpatients with dyspnoea – Area under the curve (AUC).(b) Receiver operating characteristic curve for emergencydepartment (ED)-probability, MRproANP, and both to diag-nose heart failure in patients with dyspnoea – ED-probabilityalone compared to the combination of ED-probability withMRproANP (P = 0.016).
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
124 ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129
MRproANP and NTproBNP seemed to differ regard-
ing their association with BMI (Fig. 3). Mean
MRproANP levels were not different in overweight
(BMI ‡ 30, n = 47) and obese (25 £ BMI < 30,
n = 38) patients compared to individuals with normal
BMI (BMI < 25, n = 69) (P = 0.346 for both), where-
as NTproBNP levels were lower in overweight and
obese patients than in normal weight patients (P =
0.025 for both). There was no statistically significant
difference for MRproANP and NTproBNP levels
between the BMI groups in patients without HF
(P = 0.222 and P = 0.314 respectively).
Discussion
In our cohort of unselected patients presenting to the
ED with dyspnoea, we compared the accuracy of
MRproANP measured by a novel sandwich immunoas-
say with that of NTproBNP in the diagnosis of HF. We
report four major findings. First, MRproANP levels
Table 2 Logistic regression analysis for prediction of heart failure by MR-proANP and NT-proBNP
Analyte
Optimal cut-off
concentration
Odds ratio
(95% CI) P
Diagnostic
accuracy, % False-positive, n False-negative, n
Unadjusted model
MR-proANP (pmol L)1) 206 26 (14–50) <0.001 84 22 25
NT-proBNP (pg mL)1) 1540 30 (16–56) <0.001 85 21 23
Adjusted modela
MR-proANP (pmol L)1) 206 20 (9–45) <0.001 83 25 23
NT-proBNP (pg mL)1) 1540 19 (9–42) <0.001 83 24 23
MRproANP, midregional pro-atrial natriuretic peptide; NTproBNP, N-terminal pro-B-type natriuretic peptide.aAdjusted for age, history of heart failure, prior myocardial infarction, glomerular filtration rate, haemoglobin, pulmonary rales, peripheraloedema, jugular venous distension.
Table 3 Diagnostic informationfor the clinical and biochemicaldiagnosis of HF
Optimal cut-off
concentration Sensitivity Specificity PPV NPV
Diagnostic
accuracy, %
Clinical judgement ‡80%a 62 100 1 0.69 79
MR-proANP (pmol L)1) 206 84 84 0.86 0.82 84
NT-proBNP (pg mL)1) 1560 85 85 0.87 0.83 85
Combinedb 91 84 0.86 0.89 88
PPV, positive predictive value; NPV, negative predictive value.aED-Probability for HF ‡ 80%. bClinical judgement (including BNP) combined with MR-proANP.
Table 4 Clinical utility providedby MRproANP in different sub-groups
Subgroup Subgroup criteria
MRproANP adds
to BNP
MRproANP adds
to NTproBNP
OR per
log10 P-value
OR per
log10 P-value
Grey zone BNP (n = 91) ‡100 and <500 pg mL)1 8.9 0.022 9.8 0.042
Renal dysfunction (n = 117) eGFR<60 mL min)1 5.1 0.291 10.1 0.132
Obesity (n = 61) BMI ‡30 kg m)2 20.6 0.063 24.8 0.084
eGRF, estimated glomerular filtration rate; BMI, body mass index; MRproANP, midregional pro-atrialnatriuretic peptide; NTproBNP, N-terminal pro-B-type natriuretic peptide.
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129 125
were significantly higher in patients with dyspnoea due
to HF as compared to patients with dyspnoea because
of other causes. Second, the diagnostic accuracy of
MRproANP was very high with an AUC of 0.92 and
comparable with NTproBNP. The AUC for
MRproANP in this study was very similar to the AUC
reported for BNP and NTproBNP in previous ED stud-
ies recruiting patients with dyspnoea [5–7, 18]. A
recent meta-analysis based on prospective ED studies
documented that BNP and NTproBNP had comparable
accuracy in the diagnosis of acute heart failure [20].
Taking together the observations from this meta-
analysis and our findings, it seems appropriate to
conclude that MRproANP, BNP and NTproBNP are
equally useful as an aid for the diagnosis of HF in
patients with dyspnoea. Third, the addition of
MRproANP levels to the clinical judgement in the ED
that included chest X-ray and BNP levels, significantly
improved diagnostic accuracy for HF. This observation
was also found for NTproBNP and the question rises
whether there exists an additional value to BNP or the
information obtained by BNP was overruled by the
clinician. Indeed, the diminished diagnostic accuracy of
the clinical judgement including BNP was 79.4%
instead of 84% for BNP alone at a cut-off value of
500pg mL)1. This clearly indicates that the clinician
overruled the information obtained by BNP and another
second biomarker either MRproANP or NTproBNP
could encourage the physician to reassess his diagnosis.
Fourth, MRproANP seems to provide incremental
diagnostic information in patients presenting with BNP
levels in the grey zone.
This study has particular strengths. First, we enrolled
unselected, elderly patients presenting with dyspnoea
and which are therefore representative for an ED
population presenting with dyspnoea and diagnostic
uncertainty [6, 21]. Second, this is the first large
observational cohort study assessing patients with
dyspnoea that included BNP levels in the clinical
judgement in the ED and the adjudicated gold
standard diagnosis. We consider this aspect to be of
major importance as BNP levels have previously
been shown to significantly improve diagnostic
accuracy in the ED [6]. Accordingly, our adjudi-
cated gold standard diagnosis can be considered
superior to the gold standard diagnosis in previous
studies.
The evaluation and management of patients present-
ing to the ED with dyspnoea is challenging [4, 22,
23]. The currently recommended approach to patients
with dyspnoea in the ED is based on the combina-
tion of clinical evaluation, ECG, chest X-ray and
natriuretic peptides [5, 6, 24]. Our findings validate
and extend the observations made in a recent pilot
study evaluating MRproANP for the diagnosis of HF
[25]. Gegenhuber et al. retrospectively compared
MRproANP levels with NTproBNP and BNP levels
in 251 patients presenting with dyspnoea and sug-
gested that the diagnostic information obtained by
MRproANP measurements was similar to that
obtained with either NTproBNP or BNP measure-
ments. Our study differs from theirs mainly in three
aspects: (i) we measured MRproANP and NTproBNP
at the same time-point, whereas MRproANP was
measured 1 year after the initial measurement of
BNP and NTproBNP in their study; (ii) our results
were obtained in a cohort that can be considered rep-
resentative for ED patients with dyspnoea, whilst
their results were obtained in a nearly exclusive
(>90%) male patient population; (iii) the accuracy
of our adjudicated gold standard diagnosis can be
Fig. 3 Mean NTproBNP (black bars) and MRproANP(grey bars) levels in patients with heart failure, grouped bybody mass index (BMI).
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
126 ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129
considered to be even higher as BNP levels were
available.
The particular advantage of this study is the availabil-
ity of an assessment of clinical judgement by the ED
physician that included ECG, chest X-ray and BNP
levels. The combination of MRproANP with the clini-
cal judgement in the ED increased the accuracy for a
correct diagnosis of HF to almost 88% and increased
the sensitivity from 62% to 91%, which resulted in
fewer misdiagnosed false negative patients with HF
from 56 patients to only 14 patients. The addition of
MRproANP to the clinical evaluation would have
detected one-third more patients with HF. This
improvement would lead to a more rapid and more
accurate therapy that would most likely be associated
with a reduction in morbidity, mortality and treatment
cost. This observation is supported by recent studies
showing the cost-effectiveness of natriuretic peptides
leading to the correct diagnosis and therefore a
reduced rate of hospital admission, rate of admission
to intensive care and time to discharge [26, 27]. Given
the enormous public health burden of HF [22, 23], this
achievement seems to be of paramount importance.
Previous studies described the grey zones of BNP and
NTproBNP and the difficulties to diagnose or exclude
heart failure in this population [28, 29]. MRproANP
seemed to provide incremental diagnostic information
to BNP and NTproBNP in this patient population and
may therefore help to mitigate one of the major limi-
tations associated with the clinical use of BNP and
NTproBNP. Nevertheless, MRproANP is not seen as
a substitute for further echocardiographic evaluation
in this patient population. Also, several recent studies
have documented lower BNP and NTproBNP levels
in obese compared with non-obese patients [30–32].
Therefore, lower natriuretic peptide cut-points are
needed in the diagnosis of HF in obese patients. In
this study, MRproANP levels were not different
between HF patients according to BMI, a potential
additional advantage.
Another important factor is the stability of
MRproANP, which exceeds 24 h at room temperature.
This could be of great value for general practitioners
or other physicians in private practice who need to
send blood samples to an external laboratory.
Several limitations apply to this study. First, all stud-
ies attempting to use a clinical diagnosis as gold stan-
dards for a syndrome have some limitations.
Although the cardiologists were able to review all
medical records and follow-up information, misclassi-
fication remains a possibility because HF is a clinical
diagnosis and no ideal gold standard exists. Echocar-
diography was not available in all patients. Echocar-
diographic imaging may be difficult to perform in
patients with dyspnoea or with coexisting condition
such as obesity or lung disease. Echocardiography
may also not be sensitive enough to delineate cardiac
causes of acute dyspnoea for example when severe
hypertension leads to pulmonary oedema. However,
the availability of BNP levels on top of routine clini-
cal care has consistently been shown to improve the
diagnostic accuracy in patients presenting with dysp-
noea to the ED [5, 6], but on the other hand, the
availability of BNP for the Gold standard diagnosis
and the correlation between BNP, NTproBNP and
MRproANP is a potential source of bias. Second, the
high prevalence of HF in our study population could
favour the diagnostic performance of the biomarkers.
Therefore, the results cannot be generalized to other
patient populations (e.g. primary care patients). Third,
the number of patients in the grey zone for BNP or
NTproBNP was too small to draw definite conclu-
sions regarding the additive value of MRproANP in
these patients. Additional studies including the inter-
national, multi-centre Biomarkers in ACute Heart
Failure (BACH) study will need to quantify the num-
ber of patients reclassified with the dual marker
approach.
In conclusion, the findings from this large study of
consecutive patients with dyspnoea indicate that
MRproANP is a highly accurate test for the diagnosis
of HF. The accuracy of MRproANP is comparable
with that of BNP and NTproBNP. Particularly in areas
of uncertainty with BNP and NTproBNP (grey zone),
MRproANP seems to be of additive diagnostic
value and should be further investigated in future
studies.
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129 127
Conflict of interest statement
The study was supported by research grants from the
Swiss National Science Foundation (PP00B-102853),
Brahms, the Department of Internal Medicine, Univer-
sity Hospital Basel, the Brandenburg Ministry of Eco-
nomics, Germany and the European Regional
Development Fund (EFRE ⁄ERDF). Dr Mueller has
received research support from Abbott, Biosite,
Brahms, Roche and Siemens as well as speaker’s hon-
oraria from Abbott, Bayer, Biosite, Brahms, Roche
and Dade Behring. Dr Morgenthaler is employer at
B.R.A.H.M.S. AG, a biotech company that developed
the midregional pro-atrial natriuretic peptide
(MRproANP) assay. Dr Bergmann is a member of the
board of directors and shareholder of B.R.A.H.M.S.
AG. Dr Bergmann and Dr Morgenthaler are inventors
of patents related to MRproANP.
References
1 Cleland JG, Swedberg K, Follath F et al. The EuroHeart Failure
survey programme – a survey on the quality of care among
patients with heart failure in Europe. Part 1: patient characteris-
tics and diagnosis. Eur Heart J 2003; 24: 442–63.
2 Mosterd A, Hoes AW. Clinical epidemiology of heart failure.
Heart 2007; 93: 1137–46.
3 Mueller C, Frana B, Rodriguez D, Laule-Kilian K, Perruchoud
AP. Emergency diagnosis of congestive heart failure: impact of
signs and symptoms. Can J Cardiol 2005; 21: 921–4.4 Wang CS, FitzGerald JM, Schulzer M, Mak E, Ayas NT. Does
this dyspneic patient in the emergency department have conges-
tive heart failure? JAMA 2005; 294: 1944–56.
5 Januzzi JL Jr, Camargo CA, Anwaruddin S et al. The N-termi-
nal Pro-BNP investigation of dyspnea in the emergency depart-
ment (PRIDE) study. Am J Cardiol 2005; 95: 948–54.6 Maisel AS, Krishnaswamy P, Nowak RM et al. Rapid measure-
ment of B-type natriuretic peptide in the emergency diagnosis
of heart failure. N Engl J Med 2002; 347: 161–7.7 Moe GW, Howlett J, Januzzi JL, Zowall H. N-terminal pro-
B-type natriuretic peptide testing improves the management of
patients with suspected acute heart failure: primary results of
the Canadian prospective randomized multicenter IMPROVE-
CHF study. Circulation 2007; 115: 3103–10.
8 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: 647–54.9 Mueller C, Laule-Kilian K, Frana B et al. The use of B-type
natriuretic peptide in the management of elderly patients with
acute dyspnoea. J Intern Med 2005; 258: 77–85.
10 Mukoyama M, Nakao K, Hosoda K et al. Brain natriuretic pep-
tide as a novel cardiac hormone in humans. Evidence for an
exquisite dual natriuretic peptide system, atrial natriuretic
peptide and brain natriuretic peptide. J Clin Invest 1991; 87:
1402–12.
11 Cowie MR, Struthers AD, Wood DA et al. Value of natriuretic
peptides in assessment of patients with possible new heart fail-
ure in primary care. Lancet 1997; 350: 1349–53.
12 Vesely DL. Atrial natriuretic peptide prohormone gene expres-
sion: hormones and diseases that upregulate its expression.
IUBMB Life 2002; 53: 153–9.13 Cappellin E, Gatti R, Spinella P et al. Plasma atrial natriuretic
peptide (ANP) fragments proANP (1–30) and proANP (31–67)
measurements in chronic heart failure: a useful index for heart
transplantation? Clin Chim Acta 2001; 310: 49–52.14 Ala-Kopsala M, Magga J, Peuhkurinen K et al. Molecular het-
erogeneity has a major impact on the measurement of circulat-
ing N-terminal fragments of A- and B-type natriuretic peptides.
Clin Chem 2004; 50: 1576–88.15 Morgenthaler NG, Struck J, Thomas B, Bergmann A. Immuno-
luminometric assay for the midregion of pro-atrial natriuretic
peptide in human plasma. Clin Chem 2004; 50: 234–6.
16 Mueller C, Breidthardt T, Laule-Kilian K, Christ M, Perruchoud
AP. The integration of BNP and NT-proBNP into clinical medi-
cine. Swiss Med Wkly 2007; 137: 4–12.
17 Hanley JA, McNeil BJ. A method of comparing the areas under
receiver operating characteristic curves derived from the same
cases. Radiology 1983; 148: 839–43.18 McCullough PA, Nowak RM, McCord J et al. B-type natriuretic
peptide and clinical judgment in emergency diagnosis of heart
failure: analysis from Breathing Not Properly (BNP) Multina-
tional Study. Circulation 2002; 106: 416–22.19 Stevens LA, Coresh J, Greene T, Levey AS. Assessing kidney
function – measured and estimated glomerular filtration rate.
N Engl J Med 2006; 354: 2473–83.
20 Clerico A, Fontana M, Zyw L, Passino C, Emdin M. Compari-
son of the diagnostic accuracy of brain natriuretic peptide
(BNP) and the N-terminal part of the propeptide of BNP immu-
noassays in chronic and acute heart failure: a systematic review.
Clin Chem 2007; 53: 813–22.21 Lainchbury JG, Campbell E, Frampton CM, Yandle TG, Nicho-
lls MG, Richards AM. Brain natriuretic peptide and n-terminal
brain natriuretic peptide in the diagnosis of heart failure in
patients with acute shortness of breath. J Am Coll Cardiol
2003; 42: 728–35.
22 Hunt SA, Abraham WT, Chin MH et al. ACC ⁄AHA 2005
Guideline Update for the Diagnosis and Management of
Chronic Heart Failure in the Adult: a report of the American
College of Cardiology ⁄American Heart Association Task Force
on Practice Guidelines (Writing Committee to Update the 2001
Guidelines for the Evaluation and Management of Heart Fail-
ure): developed in collaboration with the American College of
Chest Physicians and the International Society for Heart and
Lung Transplantation: endorsed by the Heart Rhythm Society.
Circulation 2005; 112: e154–235.
23 Dickstein K, Cohen-Solal A, Filippatos G et al. ESC Guidelines
for the diagnosis and treatment of acute and chronic heart fail-
ure 2008: the Task Force for the Diagnosis and Treatment of
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
128 ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129
Acute and Chronic Heart Failure 2008 of the European Society
of Cardiology. Developed in collaboration with the Heart Fail-
ure Association of the ESC (HFA) and endorsed by the Euro-
pean Society of Intensive Care Medicine (ESICM). Eur Heart J
2008; 29: 2388–442.24 Collins SP, Ronan-Bentle S, Storrow AB. Diagnostic and prog-
nostic usefulness of natriuretic peptides in emergency depart-
ment patients with dyspnea. Ann Emerg Med 2003; 41: 532–45.
25 Gegenhuber A, Struck J, Poelz W et al. Midregional pro-A-type
natriuretic peptide measurements for diagnosis of acute destabi-
lized heart failure in short-of-breath patients: comparison with
B-type natriuretic peptide (BNP) and amino-terminal proBNP.
Clin Chem 2006; 52: 827–31.26 Mueller C, Laule-Kilian K, Schindler C et al. Cost-effectiveness
of B-type natriuretic peptide testing in patients with acute dysp-
nea. Arch Intern Med 2006; 166: 1081–7.
27 Siebert U, Januzzi JL Jr, Beinfeld MT, Cameron R, Gazelle GS.
Cost-effectiveness of using N-terminal pro-brain natriuretic
peptide to guide the diagnostic assessment and management of
dyspneic patients in the emergency department. Am J Cardiol
2006; 98: 800–5.
28 van Kimmenade RR, Pinto YM, Januzzi JL Jr. Importance and
interpretation of intermediate (gray zone) amino-terminal pro-
B-type natriuretic peptide concentrations. Am J Cardiol 2008;
101: 39–42.
29 Brenden CK, Hollander JE, Guss D et al. Gray zone BNP lev-
els in heart failure patients in the emergency department: results
from the Rapid Emergency Department Heart Failure Outpatient
Trial (REDHOT) multicenter study. Am Heart J 2006; 151:
1006–11.
30 Krauser DG, Lloyd-Jones DM, Chae CU et al. Effect of body
mass index on natriuretic peptide levels in patients with acute
congestive heart failure: a ProBNP Investigation of Dyspnea in
the Emergency Department (PRIDE) substudy. Am Heart J
2005; 149: 744–50.
31 Mehra MR, Uber PA, Park MH et al. Obesity and suppressed
B-type natriuretic peptide levels in heart failure. J Am Coll Car-
diol 2004; 43: 1590–5.32 Wang TJ, Larson MG, Levy D et al. Impact of obesity on plasma
natriuretic peptide levels. Circulation 2004; 109: 594–600.
Correspondence: Dr med. Mihael Potocki, Department of Internal
Medicine, University Hospital Basel, Petersgraben 4, 4031 Basel,
Switzerland.
(fax: ++41 ⁄ 612655353; e-mail: potockim@uhbs.ch).
M. Potocki et al. | MRproANP and NTproBNP in the diagnosis of heart failure
ª 2009 Blackwell Publishing Ltd Journal of Internal Medicine 267; 119–129 129