Author + information
- Received September 14, 1999
- Revision received February 11, 2000
- Accepted March 30, 2000
- Published online August 1, 2000.
- Brian D Lowes, MDa,* (, )
- Michael Higginbotham, MDa,
- Lawrence Petrovich, MD, FACCa,
- Marcus A DeWood, MDa,
- Mark A Greenberg, MDa,
- Peter S Rahko, MDa,
- G.William Dec, MD, FACCa,
- Thierry H LeJemtel, MDa,
- Robert L Roden, MSa,
- Margo M Schleman, MD, FACCa,
- Alastair D Robertson, PhDa,
- Richard J Gorczynski, PhDa,
- Michael R Bristow, MD, PhD, FACCa,
- for the Enoximone Study Group
- ↵*Reprint requests and correspondence: Brian D. Lowes, Heart Failure Treatment Program, University of Colorado Health Sciences Center, 4200 E. 9th Avenue, B120, Denver, Colorado 80126
This study was designed to evaluate the effects of low-dose enoximone on exercise capacity.
At higher doses the phosphodiesterase inhibitor, enoximone, has been shown to increase exercise capacity and decrease symptoms in heart failure patients but also to increase mortality. The effects of lower doses of enoximone on exercise capacity and adverse events have not been evaluated.
This is a prospective, double-blind, placebo-controlled, multicenter trial (nine U.S. centers) conducted in 105 patients with New York Heart Association class II to III, ischemic or nonischemic chronic heart failure (CHF). Patients were randomized to placebo or enoximone at 25 or 50 mg orally three times a day. Treadmill maximal exercise testing was done at baseline and after 4, 8 and 12 weeks of treatment, using a modified Naughton protocol. Patients were also evaluated for changes in quality of life and for increased arrhythmias by Holter monitoring.
By the protocol-specified method of statistical analysis (the last observation carried-forward method), enoximone at 50 mg three times a day improved exercise capacity by 117 s at 12 weeks (p = 0.003). Enoximone at 25 mg three times a day also improved exercise capacity at 12 weeks by 115 s (p = 0.013). No increases in ventricular arrhythmias were noted. There were four deaths in the placebo group and 2 and 0 deaths in the enoximone 25 mg three times a day and enoximone 50 mg three times a day groups, respectively. Effects on degree of dyspnea and patient and physician assessments of clinical status favored the enoximone groups.
Twelve weeks of treatment with low-dose enoximone improves exercise capacity in patients with CHF, without increasing adverse events.
Although effective therapy with angiotensin-converting enzyme inhibitors (ACEIs) and beta-adrenergic blocking agents is now available for patients with mild-to-moderate class II to III heart failure, therapy for individuals with more advanced heart failure is less certain. Subjects with advanced heart failure are, by definition, severely impaired functionally and usually have a poor quality of life. Standard therapy with beta-blockers (1) and even ACEIs (2) may not be well-tolerated in patients with advanced heart failure, requiring inotropic therapy or cardiac transplantation in acceptable candidates. Unfortunately, heart transplantation is not an option for the majority of patients with end stage heart disease, because of age and donor supply limitations. Additionally, treatment of mild-to-moderate heart failure patients with the beta-blockers that can be tolerated improves survival substantially but has little or no effect on exercise capacity or quality of life (3). These observations indicate that medical treatment for chronic heart failure (CHF) needs improvement.
Enoximone is an imidazolone derivative that selectively inhibits sarcoplasmic reticulum-associated type III phosphodiesterase, which is expressed at high levels in human ventricular myocardium and vasculature (4–6). Myocardial and vascular phosphodiesterase inhibition (PDEI) increases cAMP levels, respectively activating protein kinase A and G to produce positive inotropic and vasodilator responses (4–9). Enoximone’s inotropic effects are additive with beta-blockers in vitro and in vivo (4). This additive effect means that enoximone and other PDEIs can partially restore the attenuation in myocardial beta-receptor signal transduction (coupled with chronotropic and inotropic responses) that characterizes the failing human heart (9–11). Another way in which enoximone could improve exercise responses is by preferentially increasing skeletal muscle blood flow (12). In contrast with beta-blockers, enoximone improves myocardial function without an increase in myocardial oxygen demand (13), and long-term administration of enoximone is not accompanied by beta-adrenergic receptor desensitization phenomena (14) or tolerance to effects on exercise capacity (15). Thus, oral enoximone is an agent that can potentially be used to increase exercise capacity and/or quality of life in CHF, as shown in previously reported placebo-controlled trials (16,17).
Unfortunately, enoximone used at the higher doses (4–6 mg/kg/day) that were originally tested in CHF increased mortality (18). However, enoximone at doses of ≤3.0 mg/kg/day has been used successfully as a bridge to heart transplant without an apparent increase in mortality (16). The purpose of this study was to evaluate the effects of lower doses (1–2 mg/kg/day) of enoximone on exercise capacity and adverse events in a placebo-controlled setting.
This study was conducted at nine heart failure centers in the U.S. patients with symptomatic, New York Heart Association (NYHA) class II to III heart failure between the ages of 18 and 80 years of age and with left ventricular ejection fractions ≤45% were enrolled. The study was conducted with the approval of the local ethics committee and with written, informed consent signed by each patient.
This was a multicenter, placebo-controlled, double-blind trial in patients with CHF. Enoximone at 25 (E25) or 50 mg three times a day (E50) was compared with placebo in 105 patients. The primary end point was maximal exercise duration, with various measures of symptoms as secondary end points, assessed by intention-to-treat. The exercise testing protocol consisted of a multistage modified Naughton (19). At baseline, subjects had to be able to exercise for 3 min and for no more than 16 min. Baseline exercise tests were repeated every 10 to 14 days until they were within 15% of each other, at which point the last two tests were averaged to obtain the Time 0 baseline value. If after three tests the exercise duration was not within 15% of the prior two, the average of the last two tests was used as the Time 0 baseline. Patients underwent follow-up exercise tests at 4, 8 and 12 weeks after randomization.
The primary end point of this study was the effect on maximum exercise capacity as assessed by treadmill exercise time at 12 weeks analyzed by the last observation carry-forward method. Secondary end points included adverse events, quality of life and arrhythmias between the treatment and placebo groups.
Concurrent therapy with digoxin, diuretics and sublingual nitroglycerin was allowed. Patients were excluded if they were receiving vasodilators (nitrates, ACEIs, hydralazine or prazosin) or beta-blockers that could not be safely stopped. Women of childbearing potential and patients with recent (<3 months) myocardial infarction were also excluded from the study. Patients with stenotic valvular disease, restrictive or hypertrophic cardiomyopathy or uncontrolled atrial fibrillation (mean heart rate > 110) were also excluded.
After a baseline medical stabilization period of 10 to 42 days, patients were randomized to placebo or to one of the two doses of enoximone. Clinical response was assessed by patient and physician investigator’s overall evaluations of improvement (“Global Assessment Instrument”) compared with baseline, and symptoms were assessed by the investigator’s determination of the NYHA classification as well as by a four-tier (none, mild, moderate, severe) dyspnea scale. In a subset of patients, Holter monitoring was performed at baseline, four weeks and 12 weeks.
Statistical analysis of exercise data (intent-to-treat) was evaluated in two ways. One data set was drawn from all subjects who completed each time point (“time course” analysis). The second used the “last observation carried forward” (LOCF) method by imputing missing values from the last available measurement available after baseline testing. The LOCF was the protocol-specified method of primary end point analysis of the exercise data. The between-group statistical methodology used was repeated measures analysis of variance (ANOVA), with specific comparisons between groups being made by using linear contrasts (20). This method compares each dose with placebo (between-group analysis) and each time point with baseline (between-time analysis). In addition, within-group analyses by repeated-measures ANOVA were done for all groups (time-response analysis) and for all times (dose-response analysis). Finally, repeated measures ANOVA was used to assess dose-time response.
Global assessment data were analyzed by comparing the number of subjects who were rated as improved or worsened to subjects in the other two categories in the enoximone groups versus the placebo group. The Global assessment instrument has five rank ordered categories, which were collapsed into three for purposes of data analysis. Data analysis at 4, 8 and 12 weeks was as described for the time-course method for exercise data. New York Heart Association and dyspnea scale data were analyzed by determining the number of subjects who improved or worsened by ≥1 class, with the statistical methodology as for the time-course method for exercise tolerance.
Holter monitor data were analyzed by comparing changes from baseline 24-h recordings within and between the three treatment groups using nonparametric methods due to skewed distributions. The three Holter parameters analyzed were average heart rate, premature ventricular contraction (PVCs)/h and ventricular tachycardia (VT) events/h. Adverse event data were compared among the three groups by contingency table analysis using Fisher’s Exact Test.
Hospitalizations were retrospectively tabulated from adverse event data reports and compared among the three treatment groups by contingency table analysis.
Baseline demographic data
Of the 105 patients enrolled, 20, 24 and 25 patients completed the study in the placebo, E25 and E50 groups, respectively. The baseline characteristics of the patients evaluated in this study are presented in Table 1. There were no statistically significant baseline differences among the three groups.
Maximal exercise tolerance
The within-group analysis of the “time-course” data in Figure 1 indicated a significant improvement (p < 0.05) compared with baseline in the placebo group at eight weeks, the E25 group at eight and 12 weeks and the E50 group at four, eight and 12 weeks. All three groups had a progressive significant improvement in exercise duration incorporating all four time points into the within-group analysis (time-response analysis), with changes in the enoximone groups being significant at p < 0.0001 and the placebo group at p < 0.05.
In the between-group analysis, the E50 group was significantly increased at four weeks compared with the placebo group, and the E25 and E50 groups were at statistical significance at 12 weeks (respective p values of 0.050 and 0.051). On dose-response analysis, the four-week data were statistically significant (p < 0.05), and the 8- (p = 0.18) and 12-week (p = 0.057) data exhibited trends. Compared with the placebo group, the time-response relationship trended towards significance for both the E25 (p = 0.14) and E50 (p = 0.073) groups. The dose-time response analysis of all doses and time points was nearly significant in favor of increasing response with ascending dose (p = 0.079).
The protocol-specified method of analyzing the primary end point was the between group analysis of exercise duration at 12 weeks, using the LOCF technique. These data are presented in Figure 2. The E25 group had a greater increase in exercise duration (p = 0.013 for E25 versus placebo, with respective increases of 115 ± [SEM] 36 s vs. 23 ± 30 s). For the E50 group the increase in exercise duration was by 117 ± 31 s (p = 0.003 vs. placebo). Additionally, the E50 group had a greater (p < 0.05) improvement in exercise duration compared with placebo at four weeks (by 80 s, p = 0.026) and eight weeks (by 74 s, p = 0.009) on between-group analysis. At four and eight weeks the E25 group was not significantly different from placebo.
On the within-group analysis, the placebo group was significantly increased versus baseline at eight weeks (p = 0.045) but was not at four and 12 weeks. The E25 group was significantly (p < 0.05) increased versus baseline at eight and 12 weeks, and the E50 group demonstrated an increase versus baseline at 4, 8 and 12 weeks. For the time-response analysis, the placebo group was not increased (p = 0.23) over the four time points, but the E25 (p = 0.0001) and the E50 (p = 0.0001) groups experienced increases.
On the between-group analysis, the time-response relationship was statistically significant from placebo for both the E25 (p = 0.026) and E50 (p = 0.016) groups. The dose-time response was also statistically significant (p = 0.019).
As can be observed in Table 2, at four weeks more subjects in the enoximone-treated groups were self-assessed as having improved (77% in the E25 and 70% in the E50 groups vs. 48% in the placebo group, respective p values of 0.019 and 0.055). At week 12 a similar percentage of enoximone-treated subjects (76%) continued to assess their status as improved versus 60% of the placebo patients, which was not statistically significant. In terms of subjects who considered themselves to have worsened, there were no differences between the enoximone- (10% to 15%) and placebo-treated (8% to 18%) subjects at any time point. There were no statistically significant differences between the placebo and enoximone groups in the Physician’s evaluation by Global Assessment (Table 2).
NYHA class, dyspnea scale and diuretic use
There were no significant changes in NYHA class throughout the study (Table 3). On the dyspnea scale compared with placebo, the E25 group contained more subjects with improvement or less worsening at four weeks (p = 0.038) and eight weeks (p = 0.047) with a trend at 12 weeks (p = 0.147) but only in subjects rated at baseline as having no or mild dyspnea. The same was true for the E50 group (in subjects with no or mild dyspnea at baseline vs. placebo p = 0.038 at four weeks, 0.028 at eight weeks and 0.091 at 12 weeks). In subjects who were rated as having moderate or severe dyspnea at baseline, there were no apparent differences between the placebo and either enoximone-treated group.
The improved dyspnea scale results in enoximone-treated subjects who had less dyspnea at baseline was not due to increased diuretic use, as over the course of the study eight subjects in the E25 group, seven subjects in the E50 group and seven subjects in the placebo group increased their diuretic dose by at least 40 mg of furosemide or its equivalent. There was also no difference in the number of subjects decreasing diuretic dose (two subjects on placebo, one on 25 mg three times a day enoximone and three on 50 mg three times a day enoximone).
Holter monitoring data available on a subset of subjects are presented in Table 4. Missing data at baseline or on follow-up plus the relatively high dropout rate led to only 29% of subjects in the placebo group having Holter data available at baseline and at 12 weeks. The active treatment groups had heart rate data available on 42% (E25) and 43% (E50) of subjects at baseline and at the end of the study. For heart rate the only within-group change compared with baseline was a reduction in heart rate in the placebo group at weeks 4 and 12. There were no statistically significant changes in heart rate in either enoximone group, with the E25 group tending to have a decrease and the E50 group being unchanged at 12 weeks. Note that the baseline heart rates tended to be higher in the placebo group than in the enoximone groups (p = 0.15), which contributes to a significant between-group change between the placebo group and both enoximone groups at four weeks and the placebo-E50 group at 12 weeks.
There was no effect of either enoximone dose on PVCs/h or VT events/h. As assessed by the Morganroth criteria (21), proarrhythmia was present at week 4 or 12 in 35% of placebo, 45% of the E25 and 41% of the E50 groups (p = NS). However, three of four placebo and one of two E25 patients who died did not have follow-up Holters.
Adverse events (AEs)
As shown in Table 5, AEs were reported at a nearly identical rate of 1.71/subject in the placebo group, 1.67/subject in the E25 group and 1.76/subject in the E50 group. Both enoximone-treated groups tended to have fewer numbers of subjects reporting dizziness, vertigo or hypotension as an AE (p = 0.05 for the combined enoximone groups vs. placebo).
Serious AEs were reported in six, three and three patients in the placebo, E25 and E50 groups, respectively. In addition, three placebo patients and two E25 patients died during or within a day of completing the study. A fourth placebo patient was discontinued in the third treatment week because of increasing heart failure and died six days later. There were no deaths in the E50 group (p = 0.05 vs. placebo group). Hospitalizations were not different between groups.
The results of this trial indicate that enoximone at doses of 25 mg three times a day and 50 mg three times a day increases maximal exercise capacity compared with placebo, as assessed by the protocol-specified method of end point analysis (last observation carried-forward at 12 weeks). The favorable effects of enoximone on maximal exercise capacity appeared to be dose-related, as supported by statistically significant dose and dose-time-response analyses with the carry-forward method data set. The improvement in exercise capacity appeared to be accompanied by symptomatic improvement; both the patients’ and the investigators’ assessments of improvement showed significant effects for enoximone 50 mg three times a day at week 4, plus trends for improvement at other time points and for the 25 mg three times a day dose. Additionally, subjects treated with either 25 or 50 mg three times a day had favorable effects on a dyspnea scale provided that they were only mildly symptomatic at baseline. Enoximone was generally well tolerated, as both enoximone groups tended to have fewer numbers of subjects with severe AEs, discontinuations due to AEs and deaths compared with the placebo group. On limited Holter monitor data, there was no evidence of increased arrhythmia in the enoximone-treated patients. These data indicate that enoximone, given at doses of 25 mg three times a day and 50 mg three times a day, improves exercise capacity without increasing serious AEs in subjects with class II to III CHF treated for a 12-week period.
Comparison with previous studies
Heart failure remains a syndrome characterized by impaired exercise tolerance due to cardiac contractile and chronotropic dysfunction. Despite exercise tolerance being the cornerstone of heart failure drug development from the mid-1980s to the mid-1990s, relatively few placebo-controlled multicenter clinical trials have demonstrated improved functional capacity in the absence of evidence for increased adverse effects. The Captopril Multicenter Study (22), which was the primary basis for that compound’s approval for a heart failure indication, is the most obvious example of a successful exercise trial with an excellent safety profile. Other examples include trials with lisinopril (23) and quinapril (24). However, many trials did not detect an increase in exercise capacity by an active agent compared with placebo (25–28) or documented both an increase in exercise performance and a trend towards increased AEs (29,30) which were subsequently shown to be significant (31,32).
Despite the discouraging survival results with positive inotropic agents (32–37) intermittent or sustained infusions of outpatient inotropic therapy continues as standard therapy for patients with intractable heart failure (38). This therapy persists because of cost efficacy compared with continuous inpatient therapy because it is successful in palliation of advanced symptoms and also because physicians and patients have accepted the possible trade-off of an increase in mortality for an improvement in quality of life (39,40). Oral enoximone has previously been shown to be efficacious in weaning patients from intravenous inotropic therapy and as a bridge to cardiac transplant (14). In addition, low dose enoximone therapy does not appear to increase mortality in subjects in this study or in other previous trials (14,16,40). These results indicate that low dose oral enoximone therapy could be of benefit in inotrope-dependent patients by allowing them to be weaned off intravenous therapy and improving exercise capacity and quality of life. However, this hypothesis will have to be tested in appropriate placebo-controlled trials conducted in advanced heart failure patients.
This study is limited by its short duration, relatively small sample size and—because it was conducted in the late 1980s—the lack of background ACEI therapy. Therefore, it is possible that a trial of longer duration or on different background treatment would yield different results. However, in a recent trial of low-dose enoximone used in combination with beta-blockers, most subjects were also treated with ACEIs, and enoximone appeared to be beneficial in the presence of full neurohormonal blockade (40).
Despite these limitations this trial indicates that, at low doses, enoximone has the potential to improve functional capacity in patients with heart failure, without increasing AEs. Further trials are needed to evaluate the effects of low-dose enoximone on survival and quality of life in the setting of current standard medical therapy, particularly in subjects with more advanced heart failure.
The authors wish to thank Laurel Hunter and Frank Stewart for editorial assistance and manuscript preparation.
Enoximone Study Group Members and Institutional Affiliations
|Michael Higginbotham, MD||Duke University Medical Center|
|Lawrence Petrovich, MD||Tulane University|
|Marcus A. DeWood, MD||Deaconess Medical Center|
|Mark A. Greenberg, MD||Albert Einstein College of Medicine|
|Peter S. Rahko, MD||University of Wisconsin Medical School|
|G. William Dec, MD||Massachusetts General|
|Thierry H. LeJemtel, MD||Albert Einstein College of Medicine|
☆ This Phase II clinical trial was sponsored by Marian-Merrell Dow (MMD), now Hoechst Marian Roussel (HMR). At the time of this trial, Margo Schleman was an employee of MMD. Enoximone has since been licensed by HMR to Myogen, Inc., from whom Robert Roden and Alastair Robertson draw partial salary support. Michael Bristow is an Officer and Director of Myogen, in which he holds equity. Richard Gorczynski is an employee of Myogen.
- angiotensin-converting enzyme inhibitor
- adverse event
- analysis of variance
- chronic heart failure
- last observed carry forward
- New York Heart Association
- phosphodiesterase inhibitor
- premature ventricular contraction
- ventricular tachycardia
- Received September 14, 1999.
- Revision received February 11, 2000.
- Accepted March 30, 2000.
- American College of Cardiology
- Macdonald P.S,
- Keogh A.M,
- Aboyoun C.L,
- Lund M,
- Amor R,
- McCaffrey D.J
- Packer M,
- Lee W.H,
- Kessler P.D,
- Medina N,
- Yushak M,
- Gottlieb S.S
- ↵Bristow MR. Beta-adrenergic receptor blockade in chronic heart failure. Circulation (in press).
- Dage R.C,
- Okerholm R.A
- Jiang H,
- Colbran J.L,
- Francis S.H,
- Corbin J.D
- Herrmann H.C,
- Ruddy T.D,
- Dec G.W,
- Strauss H.W,
- Boucher C.A,
- Fifer M.A
- Colucci W.S,
- Ribeiro J.P,
- Rocco M.B,
- et al.
- Fowler M.B,
- Laser J.A,
- Hopkins G.L,
- Minobe W,
- Bristow M.R
- Bristow M.R,
- Port J.D,
- Hershberger R.E,
- Gilbert E.M,
- Feldman A.M
- Uretsky B.F,
- Jessup M,
- Konstam M.A
- ↵Weber KT, Kinasewitz GT, Janicki JS, Fishman AP. Oxygen utilization and ventilation during exercise in patients with chronic heart failure. Circulation 1982;65:1213–23.
- Winer B.J
- Captopril Multicenter Research Group
- Riegger G.A.J
- MOCHA Investigators,
- Bristow M.R,
- Gilbert E.M,
- Abraham W.T,
- et al.
- Dickstein K,
- Barvik S,
- Aarsland T
- Bucindolol Investigators,
- Bristow M.R,
- O’Connell J.B,
- Gilbert E.M,
- et al.
- Packer M,
- Narahara K.A,
- Elkayam U,
- et al.
- PROFILE Investigators and Coordinators,
- Packer M,
- Rouleau J,
- Swedburg K,
- Pitt B,
- Fisher L,
- Klepper M
- Packer M,
- Medina N,
- Yushak M
- Lubsen J,
- Just H,
- Hjalmarsson A.C,
- et al.
- Dies F,
- Krell M.J,
- Whitlow P,
- et al.
- Italian Xamoterol Multicenter Research Group
- Shakar S.F,
- Abraham W.T,
- Gilbert E.M,
- et al.