Author + information
- Received September 30, 1998
- Revision received February 24, 1999
- Accepted March 31, 1999
- Published online July 1, 1999.
- Roger M. Mills, MD, FACC∗,*,
- Thierry H. LeJemtel, MD†,
- Darlene P. Horton, MD‡,
- Chang-seng Liang, MD, PhD, FACC§,
- Roberto Lang, MD, FACC∥,
- Marc A. Silver, MD, FACC¶,
- Charles Lui, MD, FACC#,
- Kanu Chatterjee, MD, FACC∗∗,
- on Behalf of the Natrecor Study Group1
- ↵*Reprint requests and correspondence: Dr. Roger M. Mills, Desk F-15, Department of Cardiology, Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44195
The goal of this study was to further define the role of nesiritide (human b-type natriuretic peptide) in the therapy of decompensated heart failure (HF) by assessing the hemodynamic effects of three doses (0.015, 0.03 and 0.06 μg/kg/min) administered by continuous intravenous (IV) infusion over 24 h as compared with placebo.
Previous studies have shown beneficial hemodynamic, neurohormonal and renal effects of bolus dose and 6-h infusion administration of nesiritide in HF patients. Longer term safety and efficacy have not been studied.
This randomized, double-blind, placebo-controlled multicenter trial enrolled subjects with symptomatic HF and systolic dysfunction (left ventricular ejection fraction ≤35%). Central hemodynamics were assessed at baseline, during a 24-h IV infusion and for 4 h postinfusion.
One hundred three subjects with New York Heart Association class II (6%), III (61%) or IV (33%) HF were enrolled. Nesiritide produced significant reductions in pulmonary wedge pressure (27% to 39% decrease by 6 h), mean right atrial pressure and systemic vascular resistance, along with significant increases in cardiac index and stroke volume index, with no significant effect on heart rate. Beneficial effects were evident at 1 h and were sustained throughout the 24-h infusion.
The rapid and sustained beneficial hemodynamic effects of nesiritide observed in this study support its use as a first-line IV therapy for patients with symptomatic decompensated HF.
The prospect of steadily increasing numbers of heart failure (HF) patients requiring hospital admission for episodes of decompensation has stimulated a search for new therapeutic agents that can safely produce rapid hemodynamic improvement, shorten hospital stays and reduce the enormous costs associated with this syndrome (1). Human b-type natriuretic peptide (hBNP), or brain natriuretic peptide, is a 32–amino acid peptide produced primarily in the heart by the left ventricle (2). Endogenous hBNP levels are elevated in patients with HF and serve as a sensitive and specific serologic marker for left ventricular dysfunction (3), prompting speculation that hBNP may play a role in the modulation of cardiac and vascular function and fluid status in this population (4). Exogenous hBNP has been assigned the generic term nesiritide. Therapeutic doses of nesiritide may have a role in the management of HF. Previous human studies have shown that intravenous (IV) administration of nesiritide produces vasodilation, antagonism of the renin–aldosterone system, diuresis and natriuresis (5–8). When administered to HF patients, IV bolus doses and 6-h infusions of nesiritide were well tolerated and resulted in significant dose-related reductions in pulmonary capillary wedge pressure (PCWP), mean right atrial pressure (MRAP) and systemic vascular resistance (SVR), with increases in cardiac index (CI) (9–11). These observations suggest that nesiritide produces a unique combination of desirable hemodynamic, neurohormonal and renal effects in HF. To further define the role of nesiritide in the therapy of HF, this study was designed to assess the hemodynamic effects of three doses of nesiritide (0.015, 0.03 and 0.06 μg/kg/min) compared with placebo, administered as a continuous IV infusion for 24 h in patients with advanced HF. Secondary objectives included assessing the renal response to and the safety of the continuous 24-h IV infusion.
For this randomized, double-blind, placebo-controlled study, the investigational review committees of all 16 participating study sites approved the study protocol and informed consent document. Subjects with New York Heart Association class II, III or IV HF were admitted to the hospital and a pulmonary artery flotation catheter was placed at least 90 min before baseline central hemodynamic measurements were made and at least 2 h before the start of study drug. Once all eligibility criteria (Table 1)were met, subjects were randomized into one of four treatment groups: group 1, IV placebo bolus followed by a placebo infusion; group 2, nesiritide 0.25-μg/kg IV bolus followed by an infusion of 0.015 μg/kg/min; group 3, nesiritide 0.5-μg/kg IV bolus followed by an infusion of 0.03 μg/kg/min; or group 4, nesiritide 1.0-μg/kg IV bolus followed by an infusion of 0.06 μg/kg/min. Randomization was stratified by study site. The randomization assigned subjects to treatment groups 1, 2, 3 or 4 in a ratio of 1:1:1:1, using an initial block of size 4 and subsequent blocks of size 8. All of the study drug, including placebo, was prepared in identical infusion bags by study site personnel who were not associated with the subject’s clinical management to ensure blinding.
Each subject received an IV loading bolus followed by a continuous IV infusion of the study drug for 24 h. Nesiritide was supplied by Scios Inc. (Mountain View, California). Placebo was a 5% dextrose in water solution supplied by the hospital pharmacy. The infusion dose was to remain constant for 24 h. If hypotension, defined as symptomatic hypotension or systemic systolic blood pressure (SBP) less than 80 to 85 mm Hg or a decrease in PCWP below 10 mm Hg occurred, the infusion flow rate was decreased by 50%. Only one dose reduction was permitted; if hypotension or excessive PCWP reduction persisted after dose reduction, the infusion was discontinued. Any subject who required the discontinuation of study drug infusion due to worsening heart failure was considered a “treatment failure.” Dose could not be increased over the originally assigned dose, or increased after a dose reduction.
Before study drug administration, beta-adrenergic blocking agents and calcium channel blocking agents were withheld for 48 h, and vasodilators, hydralazine and angiotensin-converting enzyme inhibitors were withheld for 12 to 24 h. During study drug infusion, angiotensin-converting enzyme inhibitors, vasodilators, beta-blockers and calcium channel blockers were withheld; digoxin, diuretics and antiarrhythmics were administered as per the protocol if clinically indicated. Central hemodynamics (PCWP, MRAP, SVR, CI), systemic blood pressure and heart rate measurements were obtained at baseline, during the study drug infusion at 1, 3, 6, 10 and 24 h, and at 2 and 4 h postinfusion. Additional measurements were taken during any significant event. Cardiac index was derived from body surface area and the mean of three consecutive thermodilution measurements of cardiac output falling within 15% of each other or the mean of three out of five measurements (where the high and low values were discarded). Total fluid intake and urine output were measured for 24 h after the start of study drug. Blood was drawn for serum chemistry and hematology studies at baseline, within 24 h after discontinuation of study drug and between days 20 and 30. Blood for plasma hBNP levels was drawn 15 min preinfusion, 0.25, 0.5, 1, 3, 6 and 24 h after the start of study drug infusion and at 2, 5, 15, 30, 60, 120 and 240 min after discontinuation of the study drug. Blood was drawn for anti-hBNP antibody levels at baseline and between days 20 and 30. Four hours after discontinuation of study drug infusion, the pulmonary wedge catheter was removed if appropriate, and all previously prescribed medications could resume. Subjects were observed and discharged after follow-up examination. A follow-up telephone call was made on day 7 and day 15 for assessment of adverse events and vital status.
Data were double-key entered into the Sponsor’s clinical database and validated with computerized edit checks. All plasma hBNP and serum anti-hBNP antibody determinations were performed by a single core laboratory at Scios Inc. to ensure consistency. The analyses of hemodynamic responses were performed on the “intent to treat” population. All randomized subjects were included in this population, and subjects are summarized according to the treatment group to which they were randomly assigned. The effect of treatment group was analyzed within the framework of one-way analysis of variance. An overall comparison of the four treatment groups was made with the omnibus F test. Each nesiritide treatment group was compared with the placebo group using a pairwise contrast. Dose–response effect was examined by means of a linear contrast on treatment group using equally spaced scores. When an end point was represented as the change or percent change from baseline, a one-sample ttest was run within each treatment group to test for a nonzero mean change from baseline. This test used the within-group estimate of variance, not the pooled estimate from the analysis of variance model. All reported p values are two sided. For this report, p < 0.05 is considered statistically significant. All analysis results were obtained using SAS software (Cary, North Carolina), version 6.12, running on a Digital Equipment Corporation Alphaserver 4100 5/300.
Subject enrollment and baseline features
From 16 study sites, 103 subjects were enrolled. Figure 1is a flow diagram which accounts for subjects from randomization through the treatment period. No subjects were lost to follow-up. As shown in Table 2, there were no significant differences in demographic or baseline characteristics between treatment groups.
Figure 2shows the mean observed values for PCWP, SVR, SBP and CI in all treatment groups during the 24-h infusion and the 4-h follow-up. As early as 1 h after the initiation of study drug (the first time point assessed), PCWP was significantly reduced below baseline. Mean decreases (±SD) in PCWP from baseline at 1 h in the placebo and 0.015-, 0.03-, and 0.06-μg/kg/min nesiritide dose groups were 1.6 ± 4.8, 5.6 ± 7.6, 3.5 ± 7.6 and 8.4 ± 7.1 mm Hg, respectively (p = 0.002 [linear contrast]). At 1 h corresponding mean percent decreases in PCWP from baseline were 5 ± 20%, 17 ± 19%, 12 ± 28% and 28 ± 23% mm Hg in the placebo and three nesitiride dose groups, respectively (p = 0.001 [linear contrast]). Peak hemodynamic effects were evident by 3 to 6 h of infusion in all nesiritide groups with 27 ± 23% to 39 ± 22% reductions in PCWP from baseline. At 24 h of infusion, PCWP was still significantly decreased below baseline values in two of the three nesiritide groups, group 2 and group 4. By 4 h after discontinuation of infusion, PCWP had essentially returned to pretreatment values in all groups; mean decreases in PCWP from baseline were 2.1 ± 5.1, 3.7 ± 10.0, 0.7 ± 8.3 and 3.2 ± 9.6 mm Hg in the placebo and three nesiritide dose groups, respectively (p = 0.648 [omnibus F test]). There was no evidence for a “rebound effect,” that is, an increase of PCWP to values higher than pretreatment levels, in any of the nesiritide dose groups.
Figure 2also details the effects of nesiritide and placebo on SVR, CI and SBP, respectively. Nesiritide resulted in statistically significant reductions in SVR and SBP and increases in CI relative to baseline in all three nesiritide dose groups. Mean CI increased over baseline values in all three nesiritide dose groups throughout the 24 h of infusion and returned to baseline values within 4 h of discontinuing the infusion. For example, in the 0.015-μg/kg/min nesiritide dose group, mean CI at baseline was 1.8 ± 0.4 liters/min/m2. After 3 and 24 h of nesiritide, mean CI had increased to 2.2 ± 0.4 and 2.1 ± 0.6 liters/min/m2, respectively; within 4 h of discontinuing nesiritide, mean CI had decreased to 1.9 ± 0.4 liters/min/m2. Nesiritide also resulted in statistically significant reductions in MRAP (peak decrease of 3.8 ± 3.1 to 5.2 ± 5.1 mm Hg by 6 h; p ≤ 0.044 [pairwise contrast]) and pulmonary artery pressures (peak decrease in mean pulmonary artery pressure of 7.7 ± 10.7 to 10.5 ± 7.9 mm Hg by 6 h; p ≤ 0.001 [pairwise contrast]) compared with placebo. Peak changes in stroke volume index by 6 h were −0.9 ± 5.1, +6.1 ± 7.2, + 3.1 ± 6.9 and +6.1 ± 7.3 ml/beat/m2in the placebo and three respective nesiritide dose groups (p = 0.001, pairwise contrast for the 0.015- and the 0.06-μg/kg/min nesiritide dose groups relative to placebo). Pulmonary vascular resistance also tended to decrease during infusion of all three doses of nesiritide (peak decreases of 51 ± 157 to 68 ± 233 dyn/s/cm−5), although these changes did not reach statistical significance. Heart rate did not change appreciably during nesiritide infusion, although there was a trend toward a decrease in heart rate during infusion in the 0.015- and 0.03-μg/kg/min dose groups and an increase in the 0.06-μg/kg/min dose group. Table 3specifically addresses the hemodynamic changes evident after 3 and 24 h of infusion in those patients who continued to receive drug at 3 and at least 22 h. These data confirm a significant pharmacologic response at 3 and 24 h in all hemodynamic parameters, despite the fact that 2, 6, 8 and 12 subjects in the four groups, respectively, underwent a 50% decrease in infusion dose (all but one dose decrease in group 1 and one in group 2 were protocol-specified for excessive hemodynamic response). At 3 h of infusion, before the majority of dose reductions and treatment withdrawals had occurred, all hemodynamic parameters also show a significant dose–response relationship by linear contrast.
Urine output/diuretic use
During the 24-h treatment period, net urine output or urine sodium excretion were not greater in any of the nesiritide groups than in the placebo group. Mean net urine output in the placebo and three nesiritide ascending dose groups were 475, −91, −237 and −137 ml/24 h (p = 0.113, omnibus F test). Mean urine sodium excretion in the placebo and three nesiritide ascending dose groups was 156, 85, 106 and 147 mEq/24 h (p = 0.369, omnibus F test). Diuretics were administered during the 24-h study drug infusion in 66%, 50%, 54% and 46% of subjects in the placebo and three nesiritide ascending dose groups, respectively.
Study drug compliance
Excessive decreases in PCWP during drug infusion requiring a protocol-specified dose reduction occurred in 0 (0%), 4 (18%), 7 (27%) and 9 (35%) subjects in the placebo and three ascending dose nesiritide groups, respectively (p = 0.002 [trend test]). Hypotension led to dose reduction in an additional 1 (3%), 1 (5%), 1 (4%) and 3 (12%) subjects in the placebo and three ascending nesiritide groups, respectively. Five placebo subjects and one subject in the 0.03-μg/kg/min nesiritide dose group terminated study drug infusion prematurely due to worsening congestive HF and were therefore categorized as treatment failures (p = 0.014 [trend test]). Premature termination of study drug infusion because of hypotension or excessive decreases in PCWP was dose related, reported in 0, 0, 2, and 6 subjects in the placebo and 0.015-, 0.03- and 0.06-μg/kg/min nesiritide dose groups, respectively.
A 24-h continuous IV infusion of nesiritide was generally well tolerated. The most frequently reported adverse events during nesiritide infusion were decreases in blood pressure and excessively decreased PCWP. These events were dose related and are consistent with the pharmacologic action of the drug. Hypotension during study drug infusion was reported in 2 (7%), 1 (5%), 3 (12%) and 7 (27%) of the subjects in the placebo and 0.015-, 0.03- and 0.06-μg/kg/min nesiritide dose groups, respectively (p = 0.027 [trend test]). However, the hypotension was often asymptomatic and usually did not require intervention. Symptomatic hypotension during study drug infusion was reported in only 2 (7%), 1 (5%), 1 (4%) and 4 (15%) of the subjects in the placebo and three ascending dose groups, respectively (p = 0.346 [trend test]). Nausea during study drug infusion was reported in 0 (0%), 2 (9%), 0 (0%) and 4 (15%) of the subjects in the placebo and 0.015-, 0.03- and 0.06-μg/kg/min nesiritide dose groups, respectively (p = 0.068 [trend test]). No other adverse events appeared to be associated with nesiritide administration in a dose-related manner.
Nonsustained ventricular tachycardia was reported during study drug infusion in three subjects, all of whom were in the 0.03-μg/kg/min nesiritide dose group. During the two weeks after study drug administration, nonsustained ventricular tachycardia was reported in two placebo subjects and in three, five and zero subjects in the three ascending dose nesiritide groups, respectively. The lack of a dose-related incidence of nonsustained ventricular tachycardia during and after study drug administration suggests that these events are consistent with the expected prevalence of an underlying substrate for ventricular arrhythmias in this study population. Sustained ventricular tachycardia was not reported in any subject through day 14.
Three deaths occurred during the 15-day study follow-up period, one each in the placebo group, the 0.03-μg/kg/min nesiritide group and the 0.06-μg/kg/min nesiritide group. Two additional deaths were reported after day 15, both in the placebo group. None of these deaths was believed to be study related.
None of the subjects tested (n = 59) developed anti-hBNP antibody in the follow-up period.
Human b-type natriuretic peptide blood levels
Blood levels of hBNP achieved steady state before 3 h of infusion. Three-hour levels for group 1 (placebo) were 835 ± 989 pg/ml, and for groups, 2, 3 and 4 were 2,985 ± 1,440, 3,711 ± 2,063 and 6,456 ± 4,272 pg/ml, respectively.
This double-blind, placebo-controlled study of 103 subjects with symptomatic decompensated HF showed that a 24-h continuous IV infusion of nesiritide resulted in rapid and sustained beneficial hemodynamic effects and was well tolerated in this patient population. As early as 1 h after the start of therapy (the first time point measured in this study), as well as after 24 h of infusion, the hemodynamic effects of nesiritide include a marked reduction in preload coupled with a decrease in afterload, facilitating an increase in stroke volume and cardiac output. Hemodynamic parameters promptly returned to baseline after the infusion was discontinued after 24 h without evidence of a rebound effect.
The pharmacologic effect of hBNP at the cellular level is mediated by increases in cyclic guanosine monophosphate which lead to relaxation of vascular smooth muscle (12). Hemodynamically, these effects manifest as clinically significant balanced vasodilation, reducing both systemic vascular resistance and central venous pressure with no evidence of tolerance. This balanced vasodilator effect facilitates improved forward cardiac output at reduced filling pressures, almost certainly by redistribution of mitral regurgitant flow, as observed with other vasodilators (13,14). Reduction of systemic venous pressure may also improve left ventricular function in some advanced HF patients through ventricular interdependence (15,16). In addition, reduction in left ventricular filling pressure may improve myocardial perfusion, resulting in improvement in both systolic and diastolic function, particularly in patients with obstructive coronary disease (17).
Nesiritide is not an inotrope and is not dependent on adrenergic receptors or cyclic adenosine monophosphate for pharmacologic efficacy. Our data show no increase in either heart rate or systolic blood pressure during nesiritide infusion, indicating that the beneficial hemodynamic effects produced do not entail an increase in myocardial oxygen consumption.
Effects on diuresis
We did not observe greater urine output during the nesiritide infusion compared with placebo in this study. This is difficult to interpret given the lack of baseline urine output measurements as well as the discordant use of diuretics between treatment groups during study drug infusion. However, the lack of an increase in urine output during nesiritide infusion does conflict with the independent findings of Marcus et al. (10)and LeJemtel et al. (18, personal communication), whose studies describe a moderate increase in urine output during nesiritide infusion compared with placebo.
Clinical trial design
The design of this trial required a trade-off between assessment of therapeutic efficacy, which would have allowed more liberal dose titration, and assessment of the hemodynamic changes associated with the specific drug doses. The hemodynamic results observed most likely underestimate the sustained hemodynamic benefits of nesiritide because of dose reductions and subject withdrawals, in essence removing several patients with prompt and beneficial responses to the drug from continued treatment. The lack of a linear dose–response relationship in the observed hemodynamic responses in this intent-to-treat population might represent statistical variation due to small sample size or the effect of down-titration and withdrawal for excessive pharmacologic response. A linear dose response to increasing doses of nesiritide has been documented in other studies (10,18). Nonetheless, all three doses of nesiritide administered in this study resulted in beneficial effects on cardiac hemodynamics, such as decreases in PCWP and increases in CI.
Although the highest dose studied here (0.06 μg/kg/min) resulted in the greatest effects on most hemodynamic parameters, it was also associated with a higher incidence of adverse events (such as hypotension and nausea) than the lower two doses, and premature terminations of study drug infusion due to these adverse events were more frequent in this group. Conversely, adverse events in the two lower dose groups were infrequent and generally tolerable. This suggests that doses of nesiritide in the range of 0.015 to 0.03 μg/kg/min provide an optimal safety/efficacy profile for this clinical indication.
The rapid and sustained beneficial hemodynamic effects observed during nesiritide administration in this study support the use of this agent as a first-line intravenous therapy for patients with symptomatic, decompensated congestive HF.
The authors would like to thank the subinvestigators, study coordinators, pharmacists and nurses from the participating study sites for their contribution to the study. Principal Investigators of the Natrecor Study Group are listed alphabetically with their corresponding institution: William Abraham, MD, University of Colorado Health Sciences Center, Denver, Colorado; Kanu Chatterjee, MD, University of California San Francisco, San Francisco, California; Cara East, MD, Baylor University Medical Center; Dallas, Texas; Allen D. Johnson, MD, Scripps Clinic & Research Foundation, La Jolla, California; Walter Kao, MD, Rush-Presbyterian-St. Luke’s Medical Center, Chicago, Illinois; Stuart D. Katz, MD, Columbia-Presbyterian Medical Center, New York, New York; Marrick Kukin, MD, Mt. Sinai Medical Center, New York, New York; Roberto Lang, MD, University of Chicago Hospital, Chicago, Illinois; Thierry LeJemtel, MD, Albert Einstein College of Medicine, Montefiore Medical Center, Bronx, New York; Chang-Seng Liang, MD, PhD, FACC, University of Rochester Medical Center, Rochester, New York; Charles Lui, MD, University of Arizona, Tucson, Arizona; Roger Mills, MD, FACC, University of Florida Shands Hospital, Gainesville, Florida; Barry P. Rosenzweig, MD, New York University Medical Center, New York, New York; Marc A. Silver, MD, FACP, FACC, Loyola University Medical Center, Maywood, Illinois; and Vasant N. Udhoji, MD, FACC, West Los Angeles Veterans Administration Medical Center, Los Angeles, California.
↵1 The investigators of the Natrecor Study Group and their affiliated institutions are listed under Acknowledgments.
☆ This study was supported by a grant from Scios Inc., Mountain View, California.
This study was presented in part at the 45th Annual Scientific Session of the American College of Cardiology, 1997.
- cardiac index
- human b-type natriuretic peptide
- heart failure
- mean right atrial pressure
- pulmonary capillary wedge pressure
- systemic systolic blood pressure
- systemic vascular resistance
- Received September 30, 1998.
- Revision received February 24, 1999.
- Accepted March 31, 1999.
- American College of Cardiology
- Hosoda K.,
- Nakao K.,
- Mukoyama M.,
- et al.
- Jensen K.T.,
- Carstens J.,
- Pedersen E.B.
- Hobbs R.E.,
- Miller L.W.,
- Bott-Silverman C.,
- James K.B.,
- Rincon G.,
- Grossbard E.B.
- Marcus L.S.,
- Hart D.,
- Packer M.,
- et al.
- Louie E.K.,
- Lin S.S.,
- Reynertson S.I.,
- Brundage B.H.,
- Levitsky S.,
- Rich S.
- Strauer B.E.
- LeJemtel T.H.,
- Bourge R.C.,
- Johnson A.D.,
- Horton D.P.,
- Colucci W.S.