![]()
12
| 1 | 2005
Klinisk Biokemi i Norden
BNP or not BNP – what are the questions?
Eythor Björnsson, Dept. of Allergy, Respiratory Medicine and Sleep, Landspitali -
University Hospital, Reykjavik. E-post: eythorbj@landspitali.is
What constitutes a good clinical
marker? A biomarker is, as descri-
bed by Biomarkers Definitions
Working Group “a characteris-
tic that is objectively measured
and evaluated as an indicator of
normal biologic processes, pat-
hogenic processes, or pharmaco-
logical responses to a therapeutic
intervention (1)
For a valid biomarker the US Food and Drug
Administration, in its draft recommendations to the
industry, further demands; well-established perfor-
mance characteristics
and
a widespread agreement of
the results in the medical or scientific community.
To be attractive to the clinician a clinical marker
has to meet the additional criteria of being; rela-
tively cheap, reliable in terms of sensitivity, speci-
ficity, and diagnostic accuracy, fairly insensitive to
biological variations (age etc), having a clear “cut-
off” point to differentiate health from disease and,
most important, its measurement must be clinically
meaningful in the care of the patient.
Is B-type natriuretic peptide (BNP) such a marker?
The interest in natriuretic peptides has escalated
in recent years. A quick MEDLINE search reveals
that of approximately 2600 papers on BNP more
than 60% were published after the turn of the mil-
lennium. The potential usage of these peptides as
biological markers for heart diseases, principally
heart failure,
has fuelled this surge in literature.
Most commonly biomarkers are evaluated for four
potential clinical purposes (2):
• screening for pre-clinical disease in asymptoma-
tic persons,
• diagnosis of clinical disease in patients with
symptoms of uncertain cause,
• risk stratification in patients with clinical disease
and
• guidance in the treatment of patients with known
disease
This review aims to describe the current situation of
the clinical use of BNP in this context.
Screening asymptomatic.
The most daunting task given to any biomarker
is to prove its worth in screening for disease in
the general population. A large and recent study
evaluating BNP in this setting was the Framingham
Offspring Study.
In this study, altogether 3346 participants were
monitored regularly for the occurrence of cardio-
vascular outcomes and death during a period of 5.2
years (mean). After adjustment for cardiovascular
risk factors, each increment of 1 SD in log BNP lev-
els was associated with a stunning 27 % increase in
the risk of death and a 28 % increase in the risk of a
first cardiovascular event during this period (3).
Although remarkable these results remain to be
confirmed by independent observations. One can
safely agree with the authors in that further studies are
warranted to determine whether a finding of elevated
BNP levels in asymptomatic persons should trigger
further diagnostic evaluation of heart disease.
Other researchers have hinted that BNP could
be useful in selected patient groups that are at
particular risk of cardiovascular disease. Diabetes
is associated with an increased risk of left ventricu-
lar hypertrophy, left ventricular dysfunction and
coronary artery disease and studies have implicated
BNP as means to identify sub-clinical heart disease
in diabetics (4,5).
Another possible target group are subjects with
pulmonary diseases in which perfusion of diseased
lungs impose a strain upon the right heart. There
is now substantial evidence that BNP could have a
diagnostic role in right ventricular dysfunction and
pulmonary arterial hypertension (6).
Diagnosis of acute heart failure
Patients brought to the emergency department with
acute dyspnea may sometimes present a diagnostic
challenge. While some have argued that BNP was
