Study population. We will include into the study patients admitted to our institute for AHFS
with New York Heart Association (NYHA) class III or IV symptoms. All patients should have
evidence of symptomatic heart failure with symptoms at rest or during minimal exercise
requiring intravenous therapy.
All patients should give their written informed consent before entry into the study. Patients
with clinical or laboratory evidence of an acute coronary syndrome as well as patients with
major arrhythmias will be excluded if these conditions are likely to be the main cause of the
symptoms. In contrast, patients with aspecific electrocardiographic ST segment changes and/or
with a slight elevation of plasma troponin levels may be included as these conditions are
known to be frequently associated with AHFS and it is generally not possible to establish to
which extent they are the cause rather than the consequence of the hemodynamic
decompensation. The patients in whom a coronary revascularization procedure (either PTCA or
CABG) is planned in the next 3 months and the patients with non cardiac concomitant diseases
that may influence outcome as primary factor will be excluded from the study.
Study design. Blood samples for the assessment of NT-proBNP plasma levels will be obtained in
all the patients at at least two time intervals: pre-discharge (e.g. 24 to 72 hours before
the planned date of discharge) and at discharge (e.g. at the day of discharge). Pre-discharge
samples will be taken when the patient is clinically stable, on the same oral drugs and at
the same doses which are planned to be administered after discharge.
All patients will be randomised according to a previous randomization scheme to two groups:
control group, in which the results of NT-proBNP levels will be known only retrospectively,
at the end of the trial; intervention group, in which the values of pre-discharge NT-proBNP
plasma levels will be known by the investigator on the day after that of the blood sample.
In the intervention group, it will be possible to change medical treatment and, if needed,
prolong the hospitalization, in the case that the pre-discharge NT-proBNP levels are >3000
pg/mL. This value has been associated with a 13-fold increase in mortality and heart failure
rehospitalisation rates in patients recently admitted for AHFS. (Eur J Heart Fail. 2007 Jun
15; [Epub ahead of print] PMID: 17573240) The following changes of treatment will be
possible: increase in the dose of the diuretic, association of a different diuretic (e.g. an
aldosterone antagonist); combination of digoxin therapy; increase in the dose of the
rennin-angiotensin inhibitor; association of an angiotensin receptor blocker; association of
a nitrate; short course of i.v. diuretic, vasodilator and/or inotropic therapy.
Plasma levels of NT-proBNP will then be measured at discharge. It is our aim to use in the
intervention group a treatment modality similar to that which may be expected if NT-proBNP
levels are used to guide therapy. Therefore, the clinical investigator assessing these
patients may require to the laboratory knowledge of the discharge NT-proBNP levels and delay
patient's discharge and further adjust treatment, in case that the response to treatment
(e.g. the decrease in NT-proBNP levels) be found as insufficient.
In the intervention group, a further assessment of plasma NT-proBNP levels will be repeated 7
to 21 days after discharge. If the NT-propBNP levels will still be elevated at that time, the
investigator may further change medical treatment.
A satisfactory change in NT-ProBNP plasma levels is defined a priori as either one of the
following: a) a decrease in plasma NT-proBNP below 3000 pg/mL (a value which has been
associated with a favourable prognosis in our previous study group); b) a decrease > 30% from
the first values detected (i.e. those measured 24 to 72 hours before discharge). This
percentage is based on previous studies regarding both the prognostic value of serial changes
in NT-proBNP levels and their spontaneous week to week variability in clinically stable heart
failure patients.
Additional exams. During the hospitalization, all patients will have at least daily
assessments of their clinical conditions with medical treatment titrated to obtain clinical
stability and symptoms relief, if possible. The investigator will be unaware of the
randomization code until the day of the pre-discharge sample. No differences in baseline
clinical characteristics, hospitalization duration before the planned discharge and
in-hospital treatment are expected between the two study groups.
All the patients will undergo a standard Doppler-echocardiography and chest-X ray at least
once during the hospitalization. The 6 minutes walk distance will be measured when the
patient will be clinically stable, before discharge.
Follow-up. Each patient will undergo a clinical visit or a telephone assessment of his/her
clinical conditions at 1 and 3 months after discharge and every 6 months thereafter until the
end of the study. Unplanned ambulatory visits will be allowed based on the clinical
conditions and needs of the patient. An echocardiographic exam and the assessment of the 6
minutes walk distance will be repeated 6 months after the initial hospitalization. Standard
laboratory exams will be repeated at 1, 3 and 6 months after discharge
Outcome measures Main outcome: Incidence of unplanned cardiovascular hospitalization and
cardiac deaths at 6 months in the control and the intervention groups.
Secondary outcomes: Incidence of cardiac mortality alone; combined incidence of cardiac
deaths, cardiovascular hospitalizations and unplanned ambulatory visits; number of days spent
in hospital during the first hospitalization; number of days spent in the hospital during
follow-up; changes in the left ventricular ejection fraction, volumes and filling patterns
from the first hospitalization to after 6 months; changes in NYHA class and in 6 minutes walk
distance; changes in renal function from baseline to discharge and from discharge to the
reassessment at 1, 3 and 6 months.
