|
Stockholm, Sweden (ESC Congress 2005) -- Heart failure is a clinical syndrome which has reached epidemic proportions. Its prevalence ranges from 2% to 5% of the adult population of Western countries. Its prognosis remains poor, far worse than most of the cancers, with five-year mortality rates up to 60% among men, and 45-50% among women. Its clinical course is characterized by a progressive worsening in cardiac function and symptoms leading, finally, to a condition of advanced heart failure with high hospitalization and mortality rates. We hypothesized that, at this stage, patients need a drug which can directly improve cardiac function and, namely, the capacity of the heart to fill and then to eject blood (e.g. its diastolic and systolic function). These pharmacological agents are generally called inotropic agents (inotropism = contractility). Though the administration of an oral inotropic agent may seem straightforward in advanced heart failure, these agents have been associated with increased mortality rates in randomized trials. The likely reasons for this were an increase in cardiac arrhythmias and an excessive increase in the work of an already energy starved heart. We, however, hypothesized that the administration of the oral inotropic agent, enoximone, at doses lower than in previous trials (e.g. 25 to 50 mg tid) and to patients already on optimal treatment for heart failure (beta-blockers) would have no untoward effect on survival and would improve exercise tolerance, symptoms and clinical course. Low doses of enoximone had previously been shown to be associated with an increase in maximal exercise tolerance and with a facilitation of weaning from intravenous inotropes.
The Studies of Oral Enoximone Therapy in Advanced Heart Failure (ESSENTIAL) Trial consisted of two identical components, each enrolling approximately 900 patients. These two components, My-021 in North and South America and My-026, in Eastern and Western Europe, differed only in geographic location. The ESSENTIAL trial included patients with symptomatic advanced (NYHA Class III or IV) heart failure, an impaired cardiac function (left ventricular ejection fraction less than or equal to 30%) and at least one hospitalization (or two outpatient visits) for worsening heart failure in the previous year, and optimally treated with "beta-blockers and ACE inhibitors or ARBs. Patients were randomised 1:1 to either enoximone or placebo.
There were three primary end points: 1) a reduction in time to all-cause mortality or cardiovascular (CV) hospitalization; 2) an increase in the six-minute walk (6MW) test distance; 3) symptomatic improvement on the Patient Global Assessment (PGA) questionnaire. The first end point was assessed by a pooled analysis of the My-021 and My-026 studies, while the other two end points were analysed in each trial separately. The statistical criteria for the three primary end points were: 1) mortality or CV hospitalization, p <.007 for the combined studies; 2) 6MW, p <.02 in each component study; 3) PGA, p <.02 in each study. Time to all cause mortality was also examined in a pooled analysis of the two trials as a safety end point, with the pre-specified criterion for safety being a mortality hazard ratio 95% upper confidence limit <1.30.
Efficacy follow-up of ESSENTIAL ended on December 1, 2004 after the occurrence of the pre-specified number of mortality + CV hospitalization primary end points (final number = 923). The 1854 subjects were enrolled at 211 investigative centres in 16 countries. Compared with patients enrolled in My-021 (Americas), those enrolled in My-026 (Europe) had slightly less severe heart failure, as shown by a shorter duration of symptoms, higher left ventricular ejection fraction and shorter six minute walk distance. For the entire study population no differences in baseline characteristics were present between the 928 patients who received placebo and the 926 patients who received enoximone.
Time to all cause mortality (safety end point, measured across both trials) was similar in the enoximone and placebo study groups (201 deaths out of 926 patients in the enoximone group and 210 deaths out of 928 patients in the placebo group with a relative hazard ratio of enoximone versus placebo of 0.97 (95% confidence intervals, 0.80, 1.17); p = 0.73. Similarly, time to first cardiovascular hospitalisation or all-cause mortality was similar in the two study groups (458 events out of 926 patients in the enoximone group and 465 events out of 928 patients in the placebo group with a relative hazard ratio of enoximone versus placebo of 0.98, p = 0.71). For exercise capacity measured by the 6MW distance, the median change from baseline at six months for enoximone compared to placebo was +10.0 meters (p = 0.025) in My-021 (American trial) and +1.5 meters (p = 0.82) in My-026 (European trial). This difference was influenced by a differential change from baseline in the placebo groups in the two component studies (0 meters in My-021 and 15 meters (in My-026). The 3rd co-primary end point (change in patient self assessment by PGA) did not show any difference between placebo and enoximone. The proportion of patients who reported "marked improvement" for enoximone and placebo, respectively, were 43% and 46% in My-021 (p = 0.79) and 29% and 31% in My-026 (p = 0.11).
Subgroup analyses showed an improvement in 6MW distance in the patients with more advanced myocardial dysfunction. Namely, a significant interaction was found between baseline ejection fraction and the change in 6MW distance. No difference between enoximone and placebo was observed in the patients with an ejection fraction above its median value (0.25), whereas an increase from baseline was observed in the patients with an ejection fraction below the median. In the patients with an ejection fraction below 0.25, the change from baseline in the 6 MW distance was 15 meters with enoximone and 0 meters with placebo in the combined analysis of the two trials (p=0.007). Additionally, both all-cause mortality and mortality or cardiovascular hospitalizations rates were lower with enoximone in the last one-half of follow-up (beyond 16.4 months), (5.4% with enoximone versus 8.8% with placebo, p=0.045 and 12.5% with enoximone versus 17.4% with placebo; p=0.09).
In conclusion, enoximone did not affect mortality with a slightly lower number of deaths, compared to placebo. However, it was not significantly different from placebo in any of the three co-primary endpoints for efficacy. Post-hoc analyses showed that patients with a lower LVEF (£0.25) or patients treated for >16 months may have benefited from enoximone treatment.