Patients hospitalized for treatment of decompensated heart failure (CHF) are at risk for prolonged length of stay (LOS) and frequent readmissions. Renal dysfunction and diuretic resistance contribute to this risk, particularly if renal dysfunction worsens during CHF treatment. Brain natriuretic peptide (BNP) is a hormone of myocardial cell origin with well-defined physiological effects which include arterial and venous vasodilation, suppression of adverse neurohumoral systems and favorable effects on renal hemodynamics and sodium excretion. Recombinant human BNP (Natrecor) is approved by the FDA for treatment of decompensated CHF as it has been demonstrated to lower filling pressures and improve symptoms. While clinical trials and the FDA support the use of BNP as adjuvant therapy in decompensated CHF, the extent of its efficacy in improving non-hemodynamic CHF parameters has not been fully defined.

The objective of this clinical practice protocol is to determine whether use of BNP in addition to standard therapy, will preserve renal function and facilitate diuresis in patients with CHF and mild-moderate renal impairment (creatinine clearance > 20 but < 60 ml/min) as compared to standard therapy alone. Patients admitted to the Mayo Heart Failure Service who meet entrance criteria will be randomized to standard clinical practice with or without a 48 hour infusion of BNP.

The primary endpoints will be indices of renal function and diuretic response at 1, 2 and 3 days and at discharge. Secondary endpoints will be neurohumoral function, LOS and 30-day readmission rate.

Hypothesis:

The objective of this clinical practice protocol is to determine whether use of BNP in addition to standard therapy, will preserve renal function and facilitate diuresis in patients with CHF and renal impairment as compared to standard therapy alone. The study is targeted at patients with mild to moderate renal dysfunction with adequate blood pressure who do not require inotropic therapy and are not felt to need intravenous vasodilator therapy for acute symptom control. Patients will be randomized on admission to the heart failure service and prior to initiation of therapy. Patients who receive initial therapy in the emergency department will be eligible but the therapy given in the emergency department will be recorded.

Inclusion criteria:

• Clinical diagnosis of class III-IV CHF requiring hospitalization for treatment of CHF.
• Mild - moderate renal insufficiency (20< Creatinine Clearance < 60 ml/min as calculated by the Cockcroft-Gault formula)
• Systolic BP > 90
• Stable cardiac rhythm
• Unlikely to require cardiac catheterization

Exclusion criteria:

• Inability to give informed consent
• New onset atrial fibrillation with rapid ventricular response (HR >110 bpm)
• Active ischemia
• Known or suspected stenotic valve disease
• Acute clinical need for intravenous vasodilator (including BNP) therapy (Severely symptomatic despite rest, oxygen, initial standard therapy)
• Primary reason for admission other than treatment of decompensated CHF (rhythm, device, other medical problem)

Primary Endpoints:

1. Creatinine, Creatinine Clearance at 1,2,3 days and at discharge
2. Wt loss at 1,2,3 days and at discharge
3. Fluid balance at 1,2,3 days and at discharge
4. Use of advanced therapy for diuretic resistance (inotropes, renal dopamine, ultrafiltration, dialysis)
5. Meets criteria for diuretic resistance as defined in standardized diuretic protocol

Secondary Endpoints:

1. LOS
2. 30 day readmission for CHF (hospital records ¡V Mayo and patient phone call at 30 days)
3. Plasma renin activity, aldosterone, ANP, BNP, N-proBNP, Angiotensin II, cGMP, ET, Cystatin C at baseline and just prior to end BNP infusion (BNP group) or at 48 hours after entry (standard care group).

Cockcroft-Gault Formula (Weight in Kg; Creatinine in mg/dL - Use estimated Dry Weight) Men Crt Cl = ((140-Age)*Weight) / (72*Crt) Women Crt Cl = 0.85 * ((140-Age)*Weight) / (72*Crt)

Power calculation/Sample size:

We tabulated the HF hospital service profile in regards to renal function during a three-month period from May-July 2002 where 140 patients were admitted to the CHF service. Data to calculate creatinine clearance was not available from this data base. Mild-moderate renal dysfunction was characterized by creatinine of 1.4 - 3.0 mg/dL and was present in 60 (43%) patients. This likely represents a significant underestimation of those patients eligible for the study as our patients are elderly where creatinines < 1.4 mg/dL often correlate to creatinine clearances in the 20-60 range. In the 60 patients with creatinines of 1.4-3.0 mg/dL, the mean change in creatinine during the hospitalization was +0.022 + or - 0.377 mg/dL. Using this change as that expected in the standard therapy group, enrolling 52 patients per treatment group (n=104 total) would provide 80% power (significance level of 0.05) to detect a change of 0.222 mg/dL in mean creatinine (from + 0.022 to -0.200). Using creatinine, the 0.222 mg/dL change represents an effect size of 0.58. If one allocates representative weights and ages to the creatinines at admission and discharge and calculates the mean change in creatinine clearance, a value of 1.54 + or - 6.51 ml/min is obtained. In this case, a difference of 5 ml/min in the mean change in creatinine clearance between standard therapy and treated groups would represent a somewhat higher treatment effect (0.768) indicating more than adequate power to detect a difference of 5 ml/min in the mean change in creatinine clearance between the two groups.

Randomization:

Patients will be randomized 1:1 standard therapy to BNP + standard therapy. Randomization will be stratified by creatinine clearance with two groups (20-39 ml/min and 40-59 ml/min).

Analysis:

Primary endpoints: Mean change in creatinine, creatinine clearance, fluid balance, and body weight will be calculated at 1, 2, 3 days and at discharge and compared between the BNP and standard therapy groups. Percent patients meeting criteria for diuretic resistance and percent patients receiving advanced therapy for diuretic resistance in the two groups will be compared. All data will be analyzed by intention to treat and by received active therapy for full 48 hours.

Secondary endpoints: Mean length of stay and readmission rate will be compared between the two groups. Differences between groups will be compared by Students unpaired t test for continuous variables and Fischers exact test for nominal variables. Data entry will be the duty of the nurse coordinator and will utilize both paper and an electronic case report form currently used by the CV studies unit. The Investigators will utilize the Center for Patient Oriented Research for consultation in setting up the analyses.

Recruitment potential:

> 560 patients/year admitted to CHF service If 46% have crt creatinine clearance 20-59 ml/min = 257 patients/1 year If 50% eligible patients enrolled = 129 patients/1 year If 40% eligible patients enrolled = 103 patients/1 year If 30% eligible patients enrolled = 77 patients/1 year

BNP therapy - in treatment arm only 2 ug/kg bolus and 0.01 ug/kg/min x 48 hours as long as: Systolic BP (SBP) > 90 and no symptoms thought related to decreased BP If SBP < 90 and asymptomatic- dc infusion, 250 cc NS , resume at 0.007 ug/kg/min (no bolus) once BP > 90 If SBP < 90 and symptomatic - dc infusion, 250 cc NS bolus , Trendlenberg, do not restart

Standard therapy (to be used in both arms):

• ACE inhibitors/Angiotensin Receptor Blockers - initiate or adjust per clinical judgement

• Digoxin - per clinical judgement

• Hydralazine/Isordil - per clinical judgement

• Coumadin/Aspirin - per clinical judgement

• Beta Blockers - per clinical judgement

• Calcium channel blockers - per clinical judgement

• Non-cardiac meds - per clinical judgement

• Spironolactone

  1. already on - continue at current dose or dc as appropriate
  2. already on - increase at dc if clinically indicated
  3. not on - add at dc if clinically indicated

• Other diuretic - discontinue and follow algorithm

  • Diet - 2 gm NaCl and 1.5 Liter total fluid restriction
  • Daily weights (same scale)
  • Accurate I and O
  • Oxygen - per clinical judgement

Diuretic therapy - Standardized Diuretic Algorithm based on initial creatinine clearance (see below). BNP must be started within 1-2 hours of randomization Diuretics will be started at 1 hour after start of the BNP infusion in BNP group. Diuretics can be started immediately after randomization in the standard therapy group. The study period is considered to start with initiation of BNP (BNP group) or diuretics (Standard therapy group).

Laboratory monitoring:

Standard clinical care which includes daily electrolyte panel (creatinine, blood urea nitrogen, sodium, potassium). Humoral function (BNP, ANP, PRA, Aldosterone, ET, N-proBNP, Angiotensin II, cGMP) at baseline and prior to end of Natrecor infusion (research labs). Twenty ml of blood in standard EDTA tube and five ml of blood in EDTA tube with ACE inhibitor will be needed per blood draw.

Vitals monitoring:

Standard clinical care which includes established nursing protocols for use of BNP infusion on telemetry. Introduction of BNP into clinical practice was piloted in the CCU and then on the cardiology telemetry units (4 Domatilla and 4 Joesph) by the CHF group and is now standard clinical practice on these floors.

Diuretic Algorithm:

Initial dosing and subsequent dosing of diuretics are to follow the diuretic algorithm outlined below. Can substitute bumex (1 mg bumex/ 40 mg furosemide). Can delay progression to next level if within 10% of goal fluid balance/weight. Clinician may use fluid balance OR weight to make decisions regarding progression to next level.

Inclusion Criteria:

• Clinical diagnosis of class III-IV CHF requiring hospitalization for treatment of CHF.
• Mild - moderate renal insufficiency (20< Creatinine Clearance < 60 ml/min as calculated by the Cockcroft-Gault formula)
• Systolic BP > 90
• Stable cardiac rhythm
• Unlikely to require cardiac catheterization

Exclusion Criteria:

• Inability to give informed consent
• New onset atrial fibrillation with rapid ventricular response (HR >110 bpm)
• Active ischemia
• Known or suspected stenotic valve disease
• Acute clinical need for intravenous vasodilator (including BNP) therapy (Severely symptomatic despite rest, oxygen, initial standard therapy)
• Primary reason for admission other than treatment of decompensated CHF (rhythm, device, other medical problem)