Author + information
- Received February 15, 2001
- Revision received July 18, 2001
- Accepted August 15, 2001
- Published online November 15, 2001.
- Uwe Zeymer, MD*,* (, )
- Harry Suryapranata, MD†,
- Jean Pierre Monassier, MD‡,
- Grzegorz Opolski, MD§,
- John Davies, MD∥,
- Gundars Rasmanis, MD¶,
- Gerard Linssen, MD#,
- Ulrich Tebbe, MD**,
- Rolf Schröder, MD, FACC††,
- Rolf Tiemann‡‡,
- Thomas Machnig, MD‡‡,
- Karl-Ludwig Neuhaus, MD*,
- for the ESCAMI Investigators
- ↵*Address for correspondence: Dr. Uwe Zeymer, Klinikum Kassel, Medizinische Klinik II, Mönchebergstrasse 41-43, D-34125 Kassel, Germany
We conducted an international, prospective, randomized, double-blind, placebo-controlled phase 2 trial in patients undergoing thrombolytic therapy or primary angioplasty for acute ST-elevation myocardial infarction (MI) to investigate the effect of eniporide on infarct size and clinical outcome.
Experimental studies suggest that the activity of the Na+/H+exchange (NHE) plays an important role in the unfavorable sequels of myocardial ischemia and reperfusion. Eniporide specifically inhibits the NHE-1 isoform and has been shown to limit infarct size in experimental models.
The primary efficacy end point was the infarct size measured by the cumulative release of alpha-hydroxybutyrate dehydrogenase (alpha-HDBH) (area under the curve [AUC] 0 to 72 h). In stage 1, 50, 100, 150 or 200 mg eniporide given as a 10-min infusion before start of reperfusion therapy were compared with placebo in 430 patients, and in stage 2, 100 and 150 mg eniporide were compared with placebo in 959 patients.
In stage 1, the administration of 100 mg and 150 mg eniporide resulted in smaller infarct sizes (mean alpha-HBDH AUC in U/ml × h, placebo: 44.2, 100 mg eniporide: 40.2, 150 mg eniporide: 33.9), especially in the angioplasty group. In contrast, in stage 2 there was no difference in the enzymatic infarct size between the three groups (placebo: 41.2, 100 mg eniporide: 43.0, 150 mg eniporide: 41.5). Overall there was no effect of eniporide on clinical outcome (death, cardiogenic shock, heart failure, life-threatening arrhythmias). However, there was a significant reduction of the incidence of heart failure in patients reperfused late (>4 h).
In this large study administration of the NHE-1 inhibitor eniporide, before reperfusion therapy in patients with acute ST elevation MI, did not limit infarct size or improve clinical outcome.
Infarct size is an important determinant of the short- and long-term outcome after acute myocardial infarction (AMI) (1,2). The most widely used and most effective proven
therapy to limit infarct size is early reperfusion induced by thrombolytic therapy or primary angioplasty (3,4). Although beneficial in terms of myocardial salvage, reperfusion itself may contribute to additional damage of the myocardium known as “reperfusion injury” (5–7). Various experimental studies suggest that ischemia/reperfusion injury is due, at least in part, to the Na+/H+exchange (NHE) system (8). Myocardial ischemia or reperfusion is associated with intracellular acidosis that leads to an activation of the NHE, which extrudes H+from cells in exchange for Na+. Since the ATP-dependent Na+/K+pump becomes inactive during ischemia, NHE-mediated Na+influx leads to the intracellular accumulation of Na+. This, in turn, stimulates Ca2+influx through the Na+/Ca2+exchange mechanism, leading to intracellular Ca2+overload, which mediates the unfavorable sequels of myocardial ischemia and reperfusion such as expansion of myocardial infarction (MI), myocardial stunning and arrhythmias (9). Eniporide belongs to the new class of drugs that specifically inhibit the NHE-1 isoform, which is the predominant isoform in the cardiac myocytes. Extensive preclinical studies, in vitro and in animals, have suggested that NHE inhibition with eniporide before the onset of ischemia is a very effective and reproducible means of limiting the extent of infarction (10,11)and that this agent provides protective benefit even when given just before reperfusion (10).
The Evaluation of the Safety and Cardioprotective Effects of Eniporide in AMI (ESCAMI) study was conducted to investigate the hypothesis that eniporide would eventually increase myocardial salvage when given as an adjunct to reperfusion treatment of AMI.
ESCAMI was an international, prospective, randomized, double-blind, placebo-controlled phase II dose finding study. All patients gave written informed consent, and the study protocol was approved by each center’s institutional ethics committee.
Patients aged between 18 and 75 years presenting with chest pain suggesting an AMI and lasting for at least 30 min were eligible for enrollment. The presence of an ST-segment elevation of ≥0.1 mV in at least two limb leads or ≥0.2 mV in two contiguous chest leads of the same vascular territory on the admission electrocardiogram (ECG) was required for enrollment. The start of reperfusion therapy was to be initiated within 6 h after the start of symptoms. The decision to perform primary angioplasty or to treat the patient with thrombolytic therapy was left to the discretion of the treating physician. Exclusion criteria were prehospital thrombolysis, Killip class IV on admission, known history of renal failure, history of severe allergic reaction, history of autoimmune diseases, pregnancy, severe concurrent illness with reduced short-term prognosis, inability to give informed consent and participation in another study within the past 30 days.
The blinded study medication was administered with a syringe pump over 10 min in a separate intravenous line. In patients receiving thrombolytic therapy, the infusion had to be completed at least 15 min after the start of treatment, while, in patients treated with primary angioplasty, the infusion had to be completed at least 10 min before the start of percutaneous transluminal coronary angioplasty (PTCA).
Study end points
The primary efficacy end point was infarct size determined by the cumulative release of alpha-hydroxybutyrate dehydrogenase (alpha-HBDH) (area under the curve [AUC]: 0 to 72 h) (12). Secondary cardiac markers, including creatine kinase (CK), CK-isoenzyme (CK-MB), troponin T and I were examined. Other secondary end points were clinical events occurring within the first six weeks, in particular, death, sustained ventricular arrhythmias, resuscitation from cardiac arrest, cardiogenic shock, heart failure, major bleedings, stroke, need for revascularization, recurrent ischemia, reinfarctions and rehospitalization. In addition, an analysis of the extent of ST-segment resolution, a marker of the actual perfusion status of the myocardium, was done.
The ESCAMI study employed a two-stage adaptive design (13,14). In stage 1, four doses were considered: 50 mg, 100 mg, 150 mg and 200 mg eniporide (Merck KGaA, Darmstadt, Germany). The objectives of stage 1 were threefold: to obtain some initial evidence of the primary efficacy end point, to select a subset of doses to be carried forward into stage 2 and to determine the number of patients to be recruited for stage 2. A monotone dose-response relation was assumed, and, thus, a one-sided linear trend test was employed to test for efficacy at the first stage. A one-sided formulation of the null hypothesis was adopted to avoid conflicting directional decisions between the two stages of the trial. The global null hypothesis was tested by means of combining the results of tests of the individual hypothesis. If the individual null hypothesis is tested in a two-sided context, problems with conflicting directional decisions could arise. Therefore, the significance level was fixed at alpha = 2.5%, one-sided. A sample size of 100 patients per group was fixed without power calculations.
For stage 2, the primary test for efficacy was again addressed comparing the mean response for the most effective dose selected from stage 1, the 150-mg dose group, with placebo. In addition, 100 mg eniporide was identified as a dose, which might also provide a clinically relevant effect, showing a 9% enzymatic infarct size reduction in stage 1. The one-sided ttest was applied for this comparison, leading to a p value, p2. The test of the global intersection null hypothesis over both stages was to be performed by combining the separate p values p1and p2by Fisher combination test. Multiple inference on the individual dose-placebo comparisons was to be performed by stepwise procedure based on the relation among the doses. If p2≤alpha, then the individual p values for the one-sided comparison of 150 mg versus placebo from both stages were to be combined by Fisher product test at full level alpha = 0.025. Since the p value for this comparison at stage 1 was considerably smaller than the overall p1value, the proof of principle would always prove the 150-mg dose to be superior to placebo. Hence, if the proof of principle had succeeded, one might immediately have proceeded to the individual comparison of 100 mg versus placebo at full level alpha. This, in analogy, was to be performed by combining the two respective one-sided p values from the two stages. In stage 1 the SD in the total sample pooled over treatments for the primary target variable AUC of alpha-HBDH was 26.17 U/ml × h. Sample size calculation yielded a sample size of 316 patients per group to be enrolled in stage 2. With 316 patients per group in stage 2 (intention-to-treat population), a true difference between two groups of Δ = 0.25 SDs could be detected for the primary target variable with a probability of 90% at alpha = 0.032 significance level (one-sided, two-sample problem, normal distribution). The power would be 88% at alpha = 0.025 significance level.
Samples of blood for assessment of alpha-HBDH, CK and CK-MB were collected on admission and after 4, 8, 12, 16, 24, 36, 48, 60 and 72 h. Troponin I and T were measured every 24 h up to day 3. At the study sites, the blood samples for the measurement of cardiac enzymes were centrifuged (1,500 gfor approximately 10 min), serum stored at −20°C and then shipped to the central laboratory by courier (BARC, NV Bioanalytic Research Corporation, Brussels, Belgium). The study protocol predefined specific rules for handling and imputation of missing values (14).
Twelve-lead ECGs were recorded at baseline, at 90 (70 to 110) min and 180 (120 to 240) min after the start of thrombolytic therapy and immediately after PTCA (5 to 30 min) and 90 (60 to 120) min after the start of PTCA. The sum of the ST-segment elevation was measured by a central core laboratory as previously described (15). Complete resolution was defined as resolution of the initial sum of ST-segment elevation ≥70%. Partial resolution was defined as ST resolution <70% to 30%, while no resolution was defined as ST resolution <30%.
Enrollment began in November 1998 and was completed in March 2000. In stage 1, a total of 433 patients were enrolled in five European countries, while in stage 2, 978 patients were enrolled in 93 centers in 10 European countries. For various reasons, 22 patients received no study medication; therefore, 1,389 patients formed our study population.
Baseline characteristics were well balanced across the treatment groups (Table 1). The mean time between start of thrombolytic therapy and start of study medication was 1.6 ± 9.7 min, and the mean time between start of study medication and start of PTCA was 27 ± 16 min. The reperfusion therapy performed in the total study population was primary PTCA in 38.0%, thrombolysis with mainly streptokinase or alteplase in 61.3% (while only 0.7% did not receive any reperfusion therapy), without any differences between the treatment groups. In patients treated with primary PTCA, the Thrombolysis in MI (TIMI) flow grade before PTCA was 0/1 in 82.4%, 2 in 9.3% and 3 in 6.8%. The intervention was successful in 91.7% of the patients; TIMI flow grade 3 was achieved in 87.7% and a TIMI flow grade 2 in 4.0% of the patients. There were no significant differences with respect to TIMI flow grade before and after PTCA between the treatment groups.
Cardiac enzyme release as a measure of infarct size
The results for the cumulative release of cardiac enzymes for stage 1 are presented in Table 2. The prespecified one-sided linear trend test revealed a p value of 0.12 for the primary end point alpha-HBDH AUC 0 to 72 h. Neither the 50- nor the 200-mg dose differed from placebo, and there was no dose-effect relationship. In the prespecified secondary analyses, statistically significant reductions of enzyme release relative to placebo were noted for alpha-HBDH, CK and CK-MB in the 100- and 150-mg eniporide groups. The effect on infarct size was most striking in the patients treated with primary PTCA (n = 165); in these patients alpha-HBDH AUC 0 to 72 h in U/ml × h was reduced from 50.7 in the placebo group to 37.7 (−25.7%) and 29.6 (−41.7%) in the 100- and 150-mg eniporide groups, respectively.
Based on these results, two doses, 100 and 150 mg, were selected for stage 2, taking into account the rules for dose selection predefined in the study protocol (14).
The positive findings of the first stage of the trial could not be confirmed in the second stage of the trial. In Table 3the results are shown and did not reveal any effect of eniporide on the primary and secondary end points. In predefined subgroups, there was no significant reduction in cardiac enzyme release by eniporide (Table 4). In another subgroup of patients with anterior infarct location treated with PTCA, no difference in the primary end point alpha-HBDH between placebo (n = 69, 50.3 ± 33.7) and patients treated with 100 mg (n = 62, 53.6 ± 31.2) or 150 mg (n = 54, 53.6 ± 36.4) eniporide was observed. Also, in patients with total occlusion before and TIMI 3 flow after PTCA, no benefit with eniporide was seen (placebo: n = 85, 44.0 ± 29.6; 100 mg: n = 78, 47.4 ± 26.7; 150 mg: n = 82, 50.8 ± 31.9).
In the total study population, the sum of ST elevations was 1.7 ± 1 mV for anterior and 0.85 ± 0.5 mV for inferior infarctions and did not differ between the treatment groups. In ECG 2 and 3, there were no differences between the placebo and the eniporide-treated patients with respect to the distribution of the three ST resolution groups. The rate of patients with complete ST resolution in ECG 2 (3)was 29.3% (46.5%), 32.2% (45.7%) and 30.7% (45.8%) in the placebo, 100-mg and 150-mg eniporide groups, respectively.
Clinical outcome and adverse events
Table 5shows the secondary end point data on the clinical outcome of all dose groups. There was no difference in clinical outcome between the 100 mg, 150 mg and placebo groups; the dose groups were evaluated in sufficient patient numbers. The overall number of deaths was low. However, there was a nonsignificant excess of deaths within six weeks (19 deaths with 100 mg, 20 with 150 mg vs. 15 with placebo) and a nonsignificant excess of stroke events in patients treated with eniporide (seven with 150 mg, three with 100 mg and one with placebo). The stroke events occurred in all but two cases (one in the placebo group and one in the 150-mg group) in patients treated with thrombolysis. Four of the seven strokes in the 150-mg group occurred within 24 h after eniporide treatment; the remaining occurred at day 4, 11 and 18, respectively; none of these strokes were due to intracranial hemorrhages.
In a subgroup of patients (n = 322) in whom reperfusion was initiated more than 4 h after symptom onset, the 150-mg dose group showed a significant reduction of patients with heart failure symptoms (Killip class ≥II) compared with placebo (placebo: 21.9 vs. eniporide: 11.1%, p = 0.02).
The overall tolerability of study medication with respect to reported adverse events (hypotension, rhythm disturbances, laboratory abnormalities or sensory symptoms) was good, and there were no significant differences between the placebo and eniporide groups.
Early reperfusion has been shown to salvage myocardium and decrease mortality in patients with evolving MI. However, reperfusion itself may induce additional cell damage. There are several approaches to protect ischemic myocardium under investigation (16). In experimental studies (10,11,17,18)and in a small clinical trial (19), inhibition of the NHE system has been shown to decrease infarct size and improve left ventricular function when given before ischemia or just before reperfusion in AMI.
Our study was the first to investigate the effects of eniporide, a specific NHE-1 inhibitor, in a large cohort of patients undergoing reperfusion therapy with either thrombolysis or primary PTCA for acute ST-elevation MI. The principal finding of this study was that a 10-min infusion of eniporide with doses up to 200 mg did not result in a reduction of infarct size measured by cardiac enzyme release. There were no subgroups of patients identified that showed a significant reduction in enzymatic infarct size. In addition, there was an apparent lack of clinical efficacy in terms of adverse cardiovascular clinical outcome. Moreover, there was a nonsignificant statistical trend towards an excess of deaths and stroke events in patients treated with eniporide. So far there were no hints from other clinical trials (19,20)with NHE inhibition for an excess in mortality. Thus, the observed trend for more deaths in patients treated with eniporide may simply have occurred by chance. The stroke rate in patients treated with thrombolysis and placebo was unexpectedly low (0.2%). Since no increase in strokes was seen in the patients treated with primary PTCA, eniporide alone did not appear to be associated with an increased risk for stroke.
Surrogate end points for infarct size
Although mortality reduction has been regarded as the standard for evaluating therapeutic efficacy of adjunctive therapies in reperfusion for AMI, surrogate end points have been advocated for a variety of reasons (21). The rather large sample size to demonstrate a significant survival is relatively prohibitive for a phase II dose-finding study. Global left ventricular ejection fraction assessed by echocardiography or angiography has been used as a surrogate end point, but there are some drawbacks for ejection fraction as surrogate end point. Angiography is an invasive procedure, is costly and is not universally available, and echocardiography has a limited technical feasibility in up to 30% of patients. Nuclear techniques are valuable to detect protective effects on the myocardium in AMI patients but are available only in a limited number of highly specialized centers. Therefore, we have chosen enzymatic infarct size as a surrogate end point (22). Cumulated alpha-HBDH release has gained clinical importance for the noninvasive estimation of infarct size because excellent correlations have been reported with anatomic infarct sizes. Due to slow elimination, alpha-HBDH seems to be superior to CK-MB in the estimation of the infarct size of patients with reperfusion therapy (12).
Causes for the lack of benefit of NHE inhibition
Various hypotheses can be proposed to explain the negative results of this study. The regimen adopted in our trial was chosen according to phase 1 studies to provide the highest tolerated dose administered as fast as possible at the time of reperfusion (23). The aim to administer the drug before reperfusion was achieved in patients treated with PTCA and thrombolytic therapy. Plasma levels of more than 3,500 ng/ml were observed within 10 min after the start of eniporide. The half-life of eniporide of approximately 2 h ensured therapeutic plasma levels of above 1,000 ng/ml even after more than 2 h (data not shown). Some of the patients treated with thrombolytic therapy might have already had recanalized vessels before the start of NHE therapy, and, in these patients, the therapeutic effect might be only minor. Still, simple technical reasons do not appear to have influenced the negative findings significantly.
Reperfusion injury and NHE inhibition
Although the phenomena of reperfusion injury are well recognized in experimental models of myocardial ischemia and reperfusion, the clinical relevance of it is not known. Thus, it may well be that reperfusion injury in the clinical setting of reperfused AMI gives no major contribution to the evolution from ischemic myocardium to myocardial necrosis (24). However, despite evidence that the NHE mechanism is active during early reperfusion and may contribute to reperfusion injury, there is only equivocal evidence from experimental studies that NHE inhibitors can protect the myocardium when they are given after coronary occlusion or immediately before reperfusion (8). In a recent experimental study, cell injury was attenuated predominantly when the NHE inhibitor cariporide was given before or during ischemia, while there was no effect when treatment was initiated at onset of reperfusion (25). These experimental findings may explain the lack of efficacy of NHE inhibition in our study. Our results do not rule out a potential benefit of NHE inhibition when given before ischemia. In a large trial with cariporide, another NHE exchange inhibitor, there was a significant clinical benefit in patients undergoing high-risk bypass surgery (20).
Our results are contrary to the findings of a small clinical trial in a highly selected group of patients with a first anterior MI who were randomized to receive placebo (n = 51) or cariporide (n = 49) immediately before reperfusion by primary PTCA (19). Significant improvements in wall motion abnormalities and a borderline significant decrease in cardiac enzyme release during the next three days were observed in the patients treated with cariporide. It is well recognized that measurements of infarct size are somewhat problematic for small clinical trials due to the very high variability of infarct size, and the numerous variables that are known to contribute to infarct size such as time to reperfusion, actual territory supplied by the infarct vessel, changes in coronary perfusion pressure and collateral blood flow are problematic as well. Particularly in small trials, these uncontrolled factors constitute a considerable risk of apparent imbalances in baseline myocardium at risk in the different treatment groups. Therefore, the positive findings of the trial with cariporide and of stage 1 of our trial might be due to chance and to the small number of patients studied, since, in the larger stage 2 of our study, no effect was seen, even in the cohort of patients with anterior MI treated with primary PTCA.
An interesting observation of the ESCAMI trial is that a subgroup of patients being reperfused late (>4 h after onset of chest pain) seemed to have a slight benefit from adjunctive eniporide in respect to enzyme release and clinical outcome, most notably in the occurrence of heart failure within one week after AMI. From this observation, one might conclude that inhibition of the cardiac myocyte NHE-1 may only be of clinical relevance if the ischemic burden has already been present for a certain period of time or that NHE-1 inhibition may have other, not yet recognized, effects besides infarct size reduction. However, since the subgroup of late reperfused patients was rather small in our trial, this finding should be further substantiated in prospectively designed clinical trials before definite conclusions can be drawn.
In conclusion, the ESCAMI trial failed to meet its primary objective to demonstrate a reduction in enzymatic infarct size and an improvement in clinical outcome in patients with ST-elevation AMI treated with adjunctive eniporide before primary PTCA or thrombolytic therapy. (Appendix 1)
Steering Committee:K.-L. Neuhaus (Chairman), J. P. Monassier, H. Suryapranata, G. Rasmanis, J. Davies
Advisory and Safety Committee:K. I. Lie (Chairman), P. Jaillon, C. Höglund
Independent Statistical Center:P. Bauer
Study Coordination and Planning:M. L. Willberg, T. Machnig, P. Verkenne, R. Euler, R. Tiemann, Merck KGaA, Darmstadt
Electrocardiogram-core Laboratory:R. Schröder
Independent Enzyme Reviewer:U. Tebbe
Central Laboratory:BARC, Bioanalytical Research Corporation, Belgium
For a complete list of participanting centers in the ESCAMI Study, please see the November 1 issue of JACCat www.cardiosource.com
☆ Supported by a grant from Merck KGaA, Darmstadt, Germany.
↵1 This manuscript is dedicated to the memory of Karl-Ludwig Neuhaus (1944–2000).
- alpha-hydroxybutyrate dehydrogenase
- acute myocardial infarction
- area under the curve
- creatine kinase
- creatine kinase isoenzyme
- Evaluation of the Safety and Cardioprotective Effects of Eniporide in Acute Myocardial Infarction study
- myocardial infarction
- percutaneous transluminal coronary angioplasty
- Received February 15, 2001.
- Revision received July 18, 2001.
- Accepted August 15, 2001.
- American College of Cardiology
- Thompson P.L,
- Fletcher E,
- Kaavatis V
- Braunwald E
- Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group
- Kloner R.A
- Matsamura K,
- Jeremy R.W,
- Schaper J,
- et al.
- Karmazyn M,
- Gan X.T,
- Humphreys R.A,
- et al.
- Guamina R.J,
- Mizumura T,
- Beier N,
- et al.
- Zeymer U,
- Suryapanata H,
- Monassier J.P,
- et al.
- Schröder R,
- Zeymer U,
- Wegscheider K,
- et al.
- Theroux P
- Klein H.H,
- Pich S,
- Bohle R.M,
- et al.
- Rohmann S,
- Weygandt H,
- Minck K.O
- Rupprecht H.J,
- vom Dahl J,
- Terres W,
- et al.
- Theroux P,
- Chaitman B.R,
- Danchin N,
- et al.
- Braunwald E,
- Cannon C.P,
- McCabe C.H
- Hermens W.T,
- van der Veen F.H,
- Willems G.M,
- et al.
- Klein H,
- Pich S,
- Bohle R.M,
- et al.