Biomarkers for diagnosis and
management of Heart Failure
BNP and NTproBNP in clinical practice
Debra Isaac BN, MD, FRCP(C)
Clinical Professor of Medicine
University of Calgary
Director – Cardiac Transplant
Foothills Medical Centre
Calgary, Alberta
2
Biomarker:
Definition and Expectation
“A characteristic that is objectively measured and evaluated as
an indicator of normal biological processes, pathogenic
processes, or pharmacologic responses to a therapeutic
intervention”
- NIH working group 2001
A cardiac biomarker can enhance clinicians’ abilities to
optimally manage patients with a cardiac disorder.
We’ve been using biomarkers to diagnosis and risk stratify
patients with cardiac ischemia for decades….
Arnold JMO, et al. Can J Cardiol 2007;23(1):21-45.
.
♥ Issues pertinent to use of BNP in Canadian Centres
9 hospital / lab administrators fear additional
costs
sts of
f adding
ddi new tests
t sts
9 cardiologists fear the BNP equivalent of
“troponitis” – and negation of cost savings
9 need for education and optimization of use of
biomarkers in HF
Approach to optimal management of CHF:
♥ First task in the management of the dyspneic patient
is correct diagnosis…. Is it heart Failure?
Sometimes obvious….sometimes not
.
Approach to optimal management of CHF:
♥ Is it heart Failure?
- SOB in patient
with COPD and known LV
-Dyspnea is a common but
dysfunction;
is it thepresentation
heart?
nonspecific
are specific
- edema, -Physical
in obesesigns
patients
(exam not always
but not sensitive for HF
easy or
straightforward)
-co-morbidities
are common in pts
presenting
with
dyspnea with
/ edema
- worsening
SOB
in patient
known LV
dysfunction but without signs of worsening
congestion
Heart Failure: A diagnostic challenge
„ Difficult diagnosis - especially in the early stage
> 60% belong to NHYA class I or II with mild or
non--conclusive symptoms
non
Fraction falsely diagnosed CHF in primary health care:
- Framingham: 40% (McKee 1971)
- Boston:
42% (Carlson 1985)
- Kuopio:
50% (Remes 1991)
1
. to optimal management of CHF:
Approach
Why do we need better ways to diagnose HF?
♥ Is it heart Failure?
‰ HF is prevalent, but may be missed or misdiagnosed, especially in its
earlier stages or in the presence of multiple coco-morbidities: Too many times
patients are treated (sometimes multiple times) for pneumonia or asthma
when the diagnosis was heart failure
CASE 1
y
with misdiagnosis:
g
‰ Added cost to health care system
Studies have shown that early diagnosis and appropriate rx of HF leads to
improved mortality and morbidity…. reduced costs if less progression and
hospitalization
‰ echocardiography not always available in timely fashion outside of
tertiary care centres
‰ echocardiography may not rule out HF just because valves work and
ventricles contract; diagnosis of HF with preserved systolic function is often
difficult!
.
♥ Is it heart Failure?
CASE 1
-BP
BP 100/70, HR 95 bpm (regular), RR 26
-JVP difficult to see (obese and bearded!)
-Chest clear, no significant edema
♥ Is it heart Failure?
CASE 2
O/E:
y working
g to breathe (RR
(
32))
-clearly
-BP 170/100, HR 98 bpm, afebrile
-JVP 4cm asa
-S4
-hyperinflated chest with reduced a/e, bibasilar
crackles, expiratory wheeze
-no edema
42 yo male presents with cough and SOB
-50 pkyr smoker, no HTN/DM, no cardiac hx
-flu-like illness started 6 weeks ago
-rx with a’biotics X 2 courses in last 4 weeks for pneumonia –
this is his 3rd clinic visit
-still SOB, fatigued, coughing (esp. at night)
-no edema, no palpitations, no chest pain
-no recent fever
Approach to optimal management of CHF:
♥ Is it heart Failure?
CASE 2
73 year old farmer presents with SOB
g smoker
-“lifelong”
-HTN, on diuretic rx, doesn’t monitor closely
-”small” MI 10 years ago
-chronic cough, increased over last month or so
-no fever or significant sputum production
-vague re; orthopnea, PND
-no edema
-no chest pain
Diagnosis of Heart Failure
CCS Recommendations:
9 Clinical hx, physical exam and laboratory testing should
be performed on all patients with suspected HF to establish
diagnosis and identify modifiable factors that may affect the
development of progression of HF
(ask yourself: is it HF, Why HF, Why now?)
2
Patients presenting to ER with SOB
Diagnosis of Heart Failure
45
CCS Recommendations:
40
35
30
25
%
20
15
10
5
9 Clinical hx, physical exam and laboratory testing should be
performed on all patients with suspected HF to establish
diagnosis and identify modifiable factors that may affect the
development of progression of HF
(ask yourself: is it HF, Why HF, Why now?)
9 Measurement of plasma B-type naturetic peptides
should be considered, where available, in patients with
suspected HF when clinical uncertainly exists
0
Hx HF
Hx COPD
HF & COPD
Harrison et al, Ann Emerg Med 2002;39:131-138
B-Type Natriuretic Peptide (BNP)
BNP and NT-proBNP
The terms BNP and
NTproBNP will be used
preproBNP (134 aa)
interchangeably during
most of this presentation
for simplicity.
p y
Synthesis and Secretion:
† Found in the cardiac ventricles
‰ BNPin
levels
may be relatively
increased and
† Released
response
to stretch
in any condition which increases
load tovolume
the ventricles:
increased
in the ventricle
Renal failure
† BNP levels Cor
arepulmonale
related to:
„
„
proBNP (108 aa)
signal peptide (26 aa)
myocyte
Increased age
Left ventricular
end-diastolic
pressure
‰ BNP levels lower
in obesity
May be initially low in flash pulmonary
NYHA‰
classification
edema
Wall stress
secretion
BNP (77-108)
NT-proBNP (1-76)
NT-proBNP is renally excreted and
biologically inactive
The active hormone that promotes
vasodilation, natriuresis and diuresis
BNP Levels in Patients With Dyspnea
Secondary to CHF or COPD
1076+/-138
1200
1076 +/- 138
1000
BNP pg/m
mL
1000
BNP pg/ml
1200
800
600
800
600
400
400
200
86 +/- 39
200
141+/-31
38+/-4
0
0
COPD
N=56
CHF
N=94
Cause of Dyspnea
Dao, Q., Maisel, A. et al. J. American College of Cardiology, Vol 37, No. 2, 2001
No CHF
N=139
LV Dysfunction
No acute CHF
N=14
CHF
N=97
Dao, Q., Maisel, A. et al. J. American College of Cardiology, Vol 37, No. 2, 2001
3
BNP Levels in Patients With Edema
Diagnosed With CHF or Without CHF
BNP in LV Dysfunction
1200
1077+/-272
1000
BNP pg
g/mL
BNP pg
g/mL
1038 +/- 163
1200
1000
800
567+/-113
600
391+/-89
400
800
600
400
63 +/- 16
200
200
30
0
0
Normal
Systolic
N=105
N=53
Diastolic
No CHF
Systolic &
Diastolic
N=42
N=44
CHF
N=44
Cause of Edema
N=14
Dao, Q., Maisel, A. et al. J. American College of Cardiology, Vol 37, No. 2, 2001
Maisel, A., De Maria, A. et al. American Heart Journal, Vol. 141, No. 3, 2001
Frequency Histogram
BNP vs. NYHA Classification
Clinical Probability of CHF (Blinded to BNP)
1500
350
1200
300
Number of Case
es
pg/m
ml
1800
900
600
300
0
Non-CHF
NYHA I
NYHA II
NT-proBNP
NYHA III
NYHA IV
Significant Indecision Exists
43%
250
200
150
100
50
BNP
0
0
10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
Roche Elecsys proBNP Product Insert 2003
Pretest Probability of CHF
Biosite Triage BNP Product Insert 2003
Indecision
n
Clarification of Diagnosis and BNP
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
BNP and NTproBNP assay cut-off points for the diagnosis of HF
BNP Reduces Clinical
Uncertainty by 74%
Age
43%
P< 0.0001
11%
Clinical
Clinical Evaluation
Evaluation
and BNP
HF
Is
unlikely
HF possible;
other dx should
be considered
HF
is
very likely
BNP
all
<100pg/ml
100 / l
100 00 / l
100-500pg/ml
>500pg/ml
00 / l
NTproBNP
<50
<300pg/ml
300-450pg/ml
>450pg/ml
NTproBNP
50-75
<300pg/ml
300-900pg/ml
>900pg/ml
NTproBNP
>75
<300pg/ml
300-1800pg/ml
>1800pg/ml
ADAPTED FROM Arnold JMO et al. Can J Cardiol 2007;23(1):21-45
4
.
♥ Is it heart Failure?
CASE 1
42 yo male presents with cough and SOB
-50
50 pkyr smoker,
smoker no HTN/DM
HTN/DM, no cardiac hx
-flu-like illness started 6 weeks ago
-rx with a’biotics X 2 courses in last 4 weeks for pneumonia –
this is his 3rd clinic visit
-still SOB, fatigued, coughing (esp. at night)
-no edema, no palpitations, no chest pain
-no recent fever
BNP: 705 pg/ml
♥ Is it heart Failure?
CASE 2
Is this HF?
1)
2)
3)
4)
Definitely yes
Possibly
Probably not
Definitely not
BNP: 200 pg/ml
73 year old farmer presents with SOB
-“lifelon
lifelong” smoker
-HTN, on diuretic rx, doesn’t monitor closely
-”small” MI 10 years ago
-chronic cough, increased over last month or so
-no fever or significant sputum production
-vague re; orthopnea, PND
-mild bilateral pre-tibial edema
-no chest pain
Diagnostic Accuracy of BNP and NT-proBNP
in pts with multiple co-morbidities presenting to Calgary EDs with SOB
1
0.9
Is this HF?
1)
2)
3)
4)
Definitely yes
Possibly
Probably not
Definitely not
-Why is BNP >100?
-How did BNP help you in this case?
Sensitivity (trrue positives)
0.8
0.7
0.6
0.5
0.4
No discrimination
0.3
NT-proBNP
AUC=0.724, p<0.0001
0.2
BNP
AUC=0.769, p<0.0001
0.1
0
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1 - Specificity (false positives)
Fig: ROC curve for detection of CHF by BNP (squares) or NT-proBNP (diamonds) in
285 patients using Diagnoses following hospital admission
5
BNP as adjunct to Clinical Assessment for diagnosis of CHF
Patient
Presentation
SOB
± peripheral edema
± orthopnea
± PND
BNP
not
indicated
NO
Caveats to BNP interpretation
†Obesity
„Levels noted to be slightly decreased1
YES
„Same is true for NT-proBNP2
Hx CHF plus evidence of
congestion including:
- elevated
ele ated JVP
- S3
- Bibasilar crackles
peripheral edema
- Reduced pulse pressure
(systolic-diastolic BP
≤30mmHg
- CXR evidence of CHF
High probability of
CHF
Hx/presentation/exam
not clearly in category
1 3,
1,
3 4 or 5
- and/or hx COPD and
CAD/CHF, increasing
shortness of breath
- and/or pt with
clinical dx COPD or
CHF who are not
responding to rx
BNP not indicated
for diagnostic
purposes
BNP indicated
for diagnosis
as adjunct to
clinical
assessment
hx COPD
no hx CHF/CAD
Pl s no eevidence
Plus
idence
of congestion on
exam or CXR
CHF unlikely
BNP not
indicated
Hx c/w
pneumonias
± fever
fe er
± elevated WBC
± CXR infiltrates
Plus: no evidence
of congestion on
exam
CHF unlikely
BNP not
indicated
Hx c/w Pulmonary
Embolism
± immobility/risk
immobilit /risk
factors for
DVT/PE
± pleuritic chest
pain
± sudden onset
SOB
±hemoptysis
± evidence of
DVT
± positive d-dimer
Consider
BNP or echo
assessment
of RV
function for
risk
stratification
†Renal function
„Levels slightly elevated but still useful as a diagnostic test3
†Pulmonary embolism/Pulmonary HTN
„Grey zone BNPs4
†Diastolic Dysfunction
„Usually less elevation than seen in systolic5
1 - Mehra et al. JACC 5/2004
2 – Krauser Am Heart 2005
3 – McCullough Am J Kidney Dis 2003
4 - Leuchte BNP in PPH JACC 2004, Kucher Circ 2003
5 - Maisel JACC 2003
BNP in HF management
Why Do We Need Better Ways to Manage HF?
♥ clinical value
9 diagnosis
™ Adjunct to (not replacement for) clinical assessment
Educated
and interpretation
is conclusive
key!
™ Most helpful
whenuse
history
and exam not
™ Potential to significantly shift probability of
diagnosis in indeterminate / unclear scenarios
† Despite evidence based Rx, HF related
mortality and morbidity remain high, and HF
related costs are high
g and increasing
g
† HF hospitalizations: high cost, frequent
readmissions – what can we do better?
† Still clinical uncertainty about when to proceed
with some of the more aggressive (and
expensive) HF therapies
Goals of Management of CHF
9 improve quality of life
9 increase life expectancy
Death by drowning
BNP-in HF Management
9 prognosis
9 therapies
Death by storm
6
BNP and Prediction of Clinical Events
¾ 72 pts admitted with adHF followed for serial BNP and 30 day events
† In patients presenting with shortness of breath
to the ED, there is a large “disconnect” between
perceived severity of CHF and the BNP level.
† Even in the setting where CHF severity is
perceived as severe, a low BNP level portends a
favorable short and long term prognosis
† Potential to guide admission from ER
BNP
+ CV events
- CV events
NO
DECREASE
15 (52%)
14 (48%)
DECREASE
7 (16%)
36 (84%)
Maisel, A et al. JACC 2004;44:1328-33
Cheung V et al. JACC 2001;37:386-391
BNP and Prediction of Clinical Events
BNP leve
elspg/mL
2000
Event-Free
1700
1400
ƒ 35 patients with previous MI,
ƒ LVEF < 35%,
35%
ƒ ICDs for primary prevention
(MADIT-II)
30-Day
R d i i
Readmission
1100
800
Death
500
200
Pre-Rx
Post-Rx
CHEST 2005; 128:2604–2610
Cheng V, et al. J Am Coll Cardiol. 2001;37:386
NT-proBNP Monitoring and Guidance
NTof HF Therapy Improves Outcomes
Cardiovascular events
NT-proBNP
Clinical
90
Event free
e (%)
Heart failure or death
90
100
100
80
70
70
60
60
50
NT-proBNP
Clinical
90
80
40
BASEL: Resources utilization
50
30
30
20
10
P = 0.049
60
90
120 150 180
Time after randomisation (days)
usual
BNP
40
40
0
70
60
50
P = 0.034
80
0
0
30
60
90
120 150 180
Hospital adm.
ICU adm. rate
30d readmission
Time after randomisation (days)
Troughton RW et al. Lancet 2000
p: 0.008
0.014
0.626Christian Müller, NEJM 2004
7
Conclusions:
BNP as a Heart Failure biomarker
BASEL results
14
12
10
8
Usual
BNP
6
4
2
BNP Testing has the potential to be a valuable
adjunct to clinical assessment in:
† Determining presence and severity of HF
† Following / assessing efficacy of therapy
† Making clinical and economic decisions in HF
patient treatment
0
t(discharge)
All p <0.01
Cost
t to Tx
Christian Müller, NEJM 2004
FYI - BC guidelines for use of BNP:
46
How to use BNP as a Heart Failure
biomarker: DIAGNOSIS
† “primary diagnosis” in difficult clinical
scenarios
† Differentiate worsening HF from other
causes of SOB in pt with known cardiac
dysfunction and other co-morbidities
How we use BNP as a Heart Failure
biomarker: MANAGEMENT/PROGNOSIS
† “dry BNP” assessment valuable in pts with
severe / recurrent HF
† Assessment of adequacy of therapy prior to
discha ge
discharge
† Determination of management plan:
9 need for admission from ER
9 need for intensive follow up / reassessment
9 need for more aggressive therapies
8
♥ clinical and economic value depends on appropriate
use of BNP measurement
9 use as adjunct (not replacement for) clinical
assessment
9 educated use and interpretation of values –
knowledge of “confounders”, optimally used as
non--binary result ((ie
non
ie not just yes or no for HF)
particularly for prognosis and management
9 use as screening for unselected population in
not recommended
DLI 2003
9