Long-Acting Angiotensin-Converting Enzyme Inhibition: Once-Daily Lisinopril Versus Twice-Daily Captopril in Mild-to-Moderate Heart Failure Roland Bach,
MD,
Once-daily liiinopril(6-20 mg) was compared with twice-daily captopril(l2.6-60 mg) in a double-blind, randomized, parallel-group study of angiotensin-converting enzyme (ACE) inhibition conducted in 31 centers for 12 weeks in patients with heart failure (New York Heart Association class IHIl) who were currently receiving digitalis and / or diuretics. The drugs were compared with regard to their effects on exercise duration, measured with bicycle ergometry, and on ectopic actii, measured u&g Holter monitoring. Both drugs significantly increased exercise duration after both 6 and 12 weeks of randomized treatment. Neither ACE inhibitor had any significant impact on the hourly rate of either ventricular ectopic counts or couplets, nor was there any difference between treatments with regard to the proportions of patients in whom ventrkular ectopic counts were reduced. Both drugs were well tolerated, with no differences observed between treatments. Potassium, urea, and creatinine levels remained stable for both treatments throughout the study. (Am J Cardiol1662;70:7OC%77C)
From the Medizinische UnversitBtsklinik, Innere Medizin III, HomburgiSaar, Germany (R.B), and Institute of Cardiology, University of Verona, Verona, Italy (P.Z.). Address for reprints: Roland Bach, MD, Medizinische Unversitltsklinik, Innere Medizin III, 6650 HomburgiSaar, Germany.
70c
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
and Piero Zardini,
F
MD
ollowing the recent publication of several major studies on the use of inhibitors of angiotensin-converting enzyme (ACE), the use of these drugs in the treatment of moderate and severe congestive heart failure (CHF; New York Heart Association [NYHA] classes III and IV) has now become standard practice as an addition to the use of diuretic drugs and/or digitalis.l This relatively new class of agents is beginning to replace the use of other vasodilators, especially as they have been shown to have superior efficacy over some of the older vasodilator therapies.2 Despite the wide acceptance of ACE inhibitors in the treatment of CHF, there remains some debate as to the relative merits of short-acting and longacting drugs of this class.3>4 Packer et al3 have demonstrated that under certain conditions, relative hypotension and renal impairment were more frequently found in patients treated with a longacting ACE inhibitor, enalapril, than in those patients treated with a short-acting ACE inhibitor, captopril. Conversely, Giles et al4 demonstrated that once-daily lisinopril is at least as effective as 3 times daily captopril when these drugs are coadministered with digitalis and/or diuretics in patients with NYHA classes II-IV heart failure. Additionally, this study demonstrated that the administration of a long-acting ACE inhibitor once daily was not associated with a significantly greater risk of adverse events. The primary aim of the study reported here was to compare the efficacy of once-daily lisinopril with twice-daily captopril in the treatment of patients with mild-to-moderate heart failure (NYHA classes II and III), maintained on optimal doses of digitalis and/or diuretics. In addition, an investigation of the effect of these ACE inhibitors on ventricular ectopic activity, as measured by 24-hour electrocardiographic Holter monitoring, was undertaken. OCTOBER
8, 1992
METHODS Patients Male or female patients > 21 years old with the clinical signs and symptoms of CHF (NYHA classes II and III), were considered eligible for entry into the study. The diagnosis of CHF was confirmed by the use of chest radiograph (cardiothoracic ratio > OS), radioisotope scan, or echocardiography (ejection fraction < 45%). All patients had to be capable of performing an exercise protocol for 4-12 minutes on a bicycle ergometer. All were symptomatic on stable doses of either digitalis or diuretics or both. Patients were excluded from the study if they had a recent history of myocardial infarction or cardiac surgery (including percutaneous transluminal coronary angioplasty), cerebrovascular accident, clinically important renal disorders, or right heart failure. Patients with intrinsic lung disease limiting exercise performance were also excluded, as were those with arrhythmias that required therapy, other than with digoxin or amiodarone. Women capable of childbearing and patients with fixed rate cardiac pacemakers were also excluded. Patients with a history of liver, blood, or endocrine disorders or clinically significant abnormal levels of serum potassium were not permitted to enter the study, as were those with any known hypersensitivities or other contraindications to ACE inhibitors. Patients with a known history of drug or alcohol abuse were also excluded, and in some centers patients who had reduced absorption following gastrointestinal surgery or those who had taken part recently in another medical study were also excluded. Patients were informed of the aims, methods, possible benefits, and hazards of the study, and gave their informed consent to enter into the study. This study was conducted in accordance with the Declaration of Helsinki (revised Venice 1983) and local ethics committee approval was obtained. Study design: The study was a double-blind, parallel-group, randomized, multicenter design. The study incorporated a placebo baseline period of lo-14 days, during which time the digoxin and/or diuretic doses were optimized according to the patients’ symptoms, and all other vasodilator and ACE inhibitor therapies were withdrawn. Patients were then randomized to either lisinopril or captopril for 12 weeks. The randomization was carried out using a separate randomization scheme, which was generated for each participating center. At the end of the baseline period, patients were assigned to a treatment group, using the computergenerated random scheme for that center. During the run-in period, placebo tablets (matching 2.5 mg
lisinopril taken once daily, and matching 12.5 mg captopril taken twice daily) were administered to the patients single blind. The initial dose of the double-blind treatment was given as a single oral dose of either 2.5 mg lisinopril or 6.25 mg captopril. Any symptoms reported or experienced by the patient were fully documented. To achieve double-blind treatment during the active treatment period, a double-dummy technique was employed. On each day patients who received lisinopril also took the appropriate number of placebo tablets matching captopril tablets and vice versa. In the case of patients randomized to lisinopril, after monitoring the effect of the initial dose (2.5 mg), 1 lisinopril tablet (5 mg) was taken per day. If, after 2 weeks of this treatment, there was evidence of clinical improvement, then the patient continued on that dosage thereafter. An increase in the dose of study medication was considered after 2 weeks of randomized treatment, and again after a further 2 weeks of randomized treatment if the following 2 safety conditions were met: (1) standing systolic blood pressure was 2 90 mm Hg, and (2) there were no symptoms of hypotension (syncope, faintness, orthostatic dizziness) on the current dose and there was evidence of a need for additional therapeutic effect in the opinion of the investigators. If the dosage of study medication needed to be increased after 2 weeks, lisinopril dosage was increased to 10 mg once daily and, again, after a further 2 weeks of treatment lisinopril could be increased to a maximum of 20 mg once daily. The dose of lisinopril was reduced at any relevant visit if (in the view of the investigator) the patient was exhibiting symptomatic hypotension or any other adverse event. For patients who received captopril, 1 captopril tablet (12.5 mg) was taken twice daily after monitoring the effect of the initial dose (6.25 mg). If after 2 weeks’ randomized treatment there was evidence of clinical improvement, then the patient continued on a dosage of captopril 12.5 mg twice daily thereafter. An increase in the dose of the study medication to 25 mg twice daily was considered after 2 weeks of randomized treatment if the previous safety conditions and the need for further efficacy were met. If after a further 2 weeks of treatment the dosage of study medication needed to be increased, captopril doses were increased to a maximum of 50 mg twice daily. As was the case for the lisinopril-treated group of patients, the dosage of captopril could also be reduced at the discretion of the investigator if the symptoms so indicated. A SYMPOSIUM:
HEART FAILURE
MANAGEMENT
71c
Concurrent medication: Patients taking digitalis were to be maintained on a constant dose throughout the study. If evidence of toxicity appeared, the dose was either reduced or treatment discontinued. The morning dose of diuretic was withheld on the day the patient received the initial dose of randomized treatment. Thereafter, the diuretic dose was adjusted during the randomizedtreatment phase as needed, based on the clinical assessment. Potassium-sparing diuretics were prohibited during the study. Serum potassium was anticipated to remain unchanged on ACE inhibitor treatment, or to increase slightly during the course of the study. However, if a patient developed hypokalemia, potassium supplements were administered. Additionally, sublingual nitroglycerin was taken as required. Chronic nitrate therapy, when used, was kept constant throughout the study. Concurrent therapy with anticoagulants was permitted; no other drugs known to affect preload or afterload could be administered during the course of this study. Clinical assessment: A thorough physical examination of each patient was conducted at the time of entry into the study. An abbreviated symptom review and physical examination was performed at randomization, and this was repeated at the end of 12 weeks of randomized treatment. Exercise testing was performed using an electrically braked bicycle ergometer and followed a schedule with continuous incremental workloads. All patients underwent a bicycle exercise test on 2 separate occasions during the placebo run-in, prior to entry into the randomized treatment phase of the study. The first exercise test served as a familiarization procedure for the patient, the second exercise test enabled the duration of exercise to be determined. Exercise tests were also carried out after 6 and 12 weeks of treatment. Patients undertook each exercise test at the same time of day (preferably in the morning), and medication was delayed on the day of the exercise test until measurements were complete. Blood pressure was measured at each clinical visit on the same arm using a mercury sphygmomanometer. Heart rate was measured on each visit by palpation of the radial pulse over a 30-second period. Recordings of blood pressure and heart rate were made after the patient had been standing for 2 minutes. Classification of the patient according to the NYHA criteria was done by the same observer throughout the study and it was assessed at the end of the placebo run-in and after 12 weeks of treatment. 72C
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
Holter monitoring was performed for a 24-hour period during placebo run-in and during the last week of treatment. The 24-hour electrocardiographic recordings were made using a 2-channel Holter monitor. Analysis of all 24-hour tapes was performed electronically by Hertford Medical (Hertford, UK). Blood samples were obtained on entry to the study and after 12 weeks of treatment for hematology and biochemistry assessments, and at weeks 2, 4, and 6 of randomized treatment, for biochemistry assessment only. Full blood count and assessment of plasma urea, sodium, potassium, and creatinine were carried out. Volunteered adverse events were obtained at each visit after the start of the placebo run-in period. In response to the question, “Have you had any symptoms or problems since your last visit?” the nature of any symptoms were recorded. Patients were asked to bring any unused tablets with them at each visit to assess compliance. Statistical analysis: Statistical analysis was directed at estimating differences between the effects of lisinopril and captopril on exercise capacity after 12 weeks of therapy. Analysis of covariante was used to compare lisinopril and captopril for the change in exercise duration from the run-in visit to treatment week 6 and the change from the run-in visit to treatment week 12. The initial analysis of covariance took account of the following sources of variation: center, covariate, treatment, and center-by-treatment interaction. The exercise duration at the run-in visit was used as the covariate. The influence of any patients withdrawing from the study or violating the protocol was investigated and the data used in the most appropriate manner to minimize bias. Previously published studies give an estimate of the variability of exercise duration data from bicycle exercise tests5-7 and indicate a between-patient standard deviation from 1.4-3.1 minutes. To detect a minimum difference of 60 seconds in exercise duration between the 2 treatment groups, it was calculated that 172 patients (86 per group) would be required. This calculation assumed a 5% level of significance with 90% power, and assumed a standard deviation of 120 seconds. To allow for withdrawals it was planned that 250 patients (125 per group) were required for this study. Three analyses of exercise duration were carried out: completers analysis, per protocol analysis, and last value analysis. Of these analyses, the completers analysis assessed the treatment effects, ignoring potential bias or loss of precision
OCTOBER
8, 1992
due to withdrawals (13% of patients in this study) and protocol violators and deviators. In this regard, it is close to an analysis based on the “intention to treat” principle. The per protocol analysis fulfilled the study objectives with respect to the protocol. The East value analysis attempted to take account of the withdrawals in estimating the treatment effects. In this article, the results of the completers analysis are reported because the results and conclusions were the same as those obtained from the per protocol
analysis.
Data on arrhythmias tend to have high variability, and the extent to which any treatment may reduce the incidence of arrhythmias is uncertain as is the clinical importance of any given reduction in CHF. Before any analysis could be undertaken, it was necessary to adjust the raw data recorded, because the individual periods of electrocardiographic monitoring were not exactly 24 hours and in many cases were much less than 24 hours. To allow for variations in the duration of electrocardiogram recording between different centers, the electrocardiographic data were converted to hourly rate for each recording, by dividing the values observed for the entire monitoring period by the duration of monitoring (hours). Only data from patients with a minimum of 8 hours of recording for both run-in and end of treatment assessment were summarized. Data from patients failing to satisfy this criterion were excluded from analyses of the electrocardiogram. In this report the results of the analysis based on patients completing the study alone are described, because the results and conclusions are the same as those obtained from a per protocol analysis. The skewed nature of the ventricular ectopic counts, couplet, and nonsustained ventricular tachycardia data suggested that it would be inappropriate to perform statistical analyses using parametric analysis of variance or covariance. These data were, therefore, analyzed using the nonparametric Wilcoxon rank-sum test. The proportion of patients in each treatment group with a reduction in the hourly rate of ventricular ectopic counts from run-in to week 12 was compared using Fisher’s exact test. The treatment effect was also assessed by summarizing the proportion of patients in each treatment group at week 12 whose ventricular ectopic activity was reduced. Two analyses of ventricular ectopic count, couplets and nonsustained ventricular tachycardia were carried out: completers analysis andperprotocol analysis, based on the principles outlined above. All safety data collected during the course of the
TABLE I Demographic Variables to Lisinopril or Captopril Variable Number patients Males (%I Females (%)
for the Patients Lisinopril
Captopril
148 117 (79) 31 (21)
108 (78) 31 (22)
Mean age (yr)
59
Age range (yr)
29-83
Body weight
Randomized
139
59
33-82 74.5
77.2
(kg)
Etiology of heart failure* (%I lschemic heart disease Cardiomyopathy Valvular heart disease Hypertension Other
52 35
49 41 8 18 5
14
22 9
*Some patients had more than 1 contributqfactorfor
heartfallure.
study were collated and the proportion of patients who reported adverse events and those who withdrew due to adverse events, were compared for both lisinopril and captopril treatments, using Fisher’s exact test. Because these analyses were concerned with treatment safety, data from all randomized patients were included. RESULTS A total of 315 patients entered the study; 28 of these withdrew during the placebo run-in period and 38 withdrew during the randomized treatment period. The demographic details of the patients randomized to treatment with lisinopril or captopril are given in Table I, which shows that both groups of patients were well matched with respect to demographic variables. At the end of the study, 43% of captopril patients were receiving 25 mg daily, 34% received 50 mg daily, and 23% were taking 100 mg daily. In the lisinopril-treated group the 5, 10, and 20 mg daily doses were being taken by 44,38, and 18% of the patients, respectively. The effects of lisinopril and captopril treatment on exercise duration for weeks 6 and 12 of randomized treatment are shown in Figure 1. At 6 weeks, the exercise duration was increased with both treatments. The increase was slightly greater for lisinopril than for captopril. However, this relatively small difference (3 seconds) was not statistically significant. Exercise duration increased by 47.2 seconds on lisinopril and 44.3 seconds on captopril (p = 0.77). Similarly, after 12 weeks of randomized treatment, exercise duration was increased by both lisinopril and by captopril. The increase seen at week 12 was slightly greater for lisinopril than for captopril, but the difference (5 seconds) was not statistically significant (p = 0.68). When compared with initial run-in values on placebo, both lisinopril and captopril treatment inA SYMPOSIUM:
HEART
FAILURE
MANAGEMENT
73c
* - * Captopril O---+ Lisinopril 600
T
_I1
(117) (117)
(115)
__---
-xc
(117) (117)
(116;
c c c r *
I
1.
I Run-in
1
I 6 weeks
I 12 weeks
FIGURE 1. The effect of liiinopril and captopril on exercise capacity at the end of the run-in period and after 6 and l2 weeks of treatment. Numbers in parentheses indicate the number of patients In each group.
creased exercise duration at both 6 and 12 weeks significantly (p = 0.0001) in all cases. 24.hour electrocardiographic monitoring: There was no statistically significant difference between the treatments for the rates of ventricular ectopic counts or couplets. The median hourly rate for ventricular tachycardia was 0. Of 90 patients who received lisinopril, 51 (57%) and 50 of 93 patients (54%) who received captopril demonstrated a reduction in the hourly rate of ventricular ectopic counts from placebo run-in to the week 12 assessment. No significant difference between the 2 treatment groups was detected (p = 0.77). Symptom review and physical examination: Figure 2 shows the percentage of randomized patients presenting with a particular symptom of 50
40
1
heart failure both at placebo run-in assessment and after 12 weeks of randomized treatment. There was an overall improvement in the monitored symptoms following treatment for both groups of patients. With regard to the percentage of patients improving, the effect of treatment was similar for both captopril and lisinopril. Mean standing blood pressures and heart rate recorded at each visit remained consistent throughout the study in both treatment groups. No effects of treatment on the mean blood pressures or heart rate were observed. Evaluation of NYHA class: Of the 125 lisinopril patients completing the study, 44 (35%) showed an improvement in NYHA status, 79 (63%) presented no change, and only 2 (1.6%) showed a deterioration in their condition. Of the 127 captopril patients completing the study, 51 (40%) showed improvement in their condition compared with 74 (58%) who showed no change and 2 (1.6%) who deteriorated (Figure 3). Adverse events: Adverse events were reported for 23 of 148 patients (16%) who received lisinopril, and 21 of 139 patients (15%) who received captopril. No significant difference in the incidence of adverse events was seen between the 2 active treatment groups. Nine of 148 patients (6%) who were randomized to lisinopril and 7 of 139 patients (5%) who were randomized to captopril withdrew from the study due to adverse events (including death). The reasons for withdrawal are shown in Table II. There were no significant differences detected between the captopril and lisinopril treatment groups. Five patients treated with lisinopril
Lisinopril
run-in
12 weeks
Lisinopril
0
Captopril
run-in
12 weeks
Captopril
m
FIGURE 2. Percentage of patients with symptoms of heart failure present at run-in and after I2 weeks of treatment with either lisinopril or captopril.
74c
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
OCTOBER
8. 1992
and 2 patients treated with captopril died during the randomized phase of the trial. The causes of death in lisinopril-treated patients were pulmonary edema, ventricular fibrillation, sudden death (n = 2), and an accident. The 2 deaths on captopril treatment were due to cardiac failure following acute myocardial infarction and pulmonary edema. None of these deaths was considered to be related to treatment with the ACE inhibitors. Three patients died during the prerandomization phase of this study due to ventricular fibrillation (n = 1) and sudden death (n = 2). A total of 15 of 148 patients (10%) who received lisinopril and 15 of 139 patients (11%) who received captopril experienced adverse events that did not lead to withdrawal from the study. No obvious treatment-related effects were observed in the laboratory parameters for either the lisinoprilor captopril-treated groups. Further, individual patient data showed no apparent treatment-related effects from either of these 2 drugs. Two patients who were treated with lisinopril showed changes in neutrophil count. For 1 of these patients, this value was above the laboratory normal range at entry and fell below the laboratory normal range after 12 weeks of treatment with lisinopril. The second patient displayed a decrease in neutrophil count but the value remained within the normal range throughout the study. Mean plasma potassium concentration was unchanged after 12 weeks of treatment with captopril, 4.33 & 0.44 mEq/liter at entry, compared with 4.35 -+ 0.49 mEq/liter after 12 weeks. In contrast, the mean plasma potassium concentration increased on lisinopril treatment from 4.30 c 0.43 mEq/liter at entry to 4.38 & 0.39 mEq/liter at the
TABLE II Withdrawal
Adverse Events Including of Treatment*
Death Leading
Lisinopril Deaths
Other
Pulmonary Accident
Captopril
edema
Pulmonary edema Cardiac failure due to acute myocardial infarction
Ventricular fibrillation Sudden death (2 patients) Abdominal pain Tingling sensation Headaches Myocardial infarction Coughing Dyspnea Blood urea nitrogen increased Hyperkalemia
events
to
Prurigo Exanthema Vertigo Feeling unwell Headaches Taste unusual Coughing Dizziness
*Some patients had more than 1 adverse event leading to withdrawal.
end of 12 weeks of treatment. Per patient, 39% of captopril-treated patients and 36% of lisinopriltreated patients showed a decrease in plasma potassium of > 0.5 mEq/liter and 53% of captopril and 56% of lisinopril patients showed an increase of plasma potassium >0.5 mEq/liter. These changes, however, were not significant, and there were no differences between treatments. Plasma urea concentrations fell slightly, but not significantly, after lisinopril and captopril treatment. At entry the mean plasma urea concentrations were 6.42 + 2.68 and 6.13 + 2.69 mmol/liter for lisinopril and captopril, respectively. After 12 weeks of treatment the plasma urea concentrations were 6.32 +- 2.48 mmol/liter for lisinopril and 5.92 5 2.35 mmol/liter for captopril. Of the captopril-treated patients, 11% showed a decrease in plasma urea concentrations greater than the predefined limit of 1.7 mmol/liter and an increase in
r
FIGURE 3. Change in functional status of patients completing l2 weeks of treatment, as assessed by New York Heart Association c[radW
Captopril
Lisinopril w
improved
c>
worsened
no change
A SYMPOSIUM:
HEART
FAILURE
MANAGEMENT
7!%
plasma urea concentration greater than 1.7 mmol/ liter was seen in 13%. The corresponding figures for the lisinopril-treated group were 13% and 14%, respectively. The mean plasma creatinine concentration was not significantly increased from 92.9 2 19.2 umol/ liter at entry to 94.0 + 20.4 pmol/liter after 12 weeks of treatment with lisinopril. Mean plasma creatinine concentration fell on captopril treatment from 93.1 & 20.2 pmol/liter at entry to 93.0 + 19.3 u,mol/liter during 12 weeks of treatment. A decrease in plasma creatinine concentration greater than the predefined limit of 17.7 pmol/liter was seen in 10% of lisinopril- and 9% of captopril-treated patients and increases in plasma creatinine >26.5 u,mol/liter were seen in 8% of both captopril- and lisinopril-treated patients. Other routine biochemical analyses indicated no obvious treatment-related effects for either lisinopril- or captopril-treated groups. The individual patient biochemical data did not show any obvious treatment-related effects. No discernible differences between the treatment groups were seen. DISCUSSION This study demonstrates the equivalence of once-daily lisinopril and twice-daily captopril in patients with mild to moderate heart failure. Both treatments produced significant, but very similar, improvements in the exercise capacity of the patients. A previous study in which once-daily lisinopril was compared with 3 times daily captopril showed a greater improvement in exercise capacity in lisinopril-treated patients compared with the captopril-treated patients (using a per protocol an~lysis).~ The results of this study appear to be at variance with this earlier work. However, there may be several reasons for the apparent lack of difference between the 2 treatments seen in this present study. This study employed bicycle ergometry to assess exercise capacity rather than a treadmill test. A previous study in which lisinopril was compared with placebo, and which used both treadmill and bicycle ergometry testing, produced different exercise times for these 2 methods with smaller differences between the ACE inhibitor and placebo for the bicycle ergometry.* However, larger differences have been observed in other placebocontrolled studies of lisinopril.9 The shorter exercise times observed using bicycle ergometry may make any small differences between treatments more difficult to detect than by using the longer exercise times of the treadmill tests. 76C
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
In the present study the time of assessment of exercise capacity in relation to the timing of the previous dose of medication was different from that reported in a previous comparative study.4 In the latter study, exercise capacity was measured during the day at no particularly fixed relation to the timing of the last dose of the study medication. However, in the present study the exercise capacity was assessed at least 24 hours after the last lisinopril dose and at least 12 hours after the last captopril dose. Under the conditions of this study, it is anticipated that the blood (and tissue) levels of both drugs would be at their respective nadirs at the time of exercise assessment, perhaps minimizing any differences between the treatments that might have been observed had blood concentrations of the respective drugs been higher. Packer,lO has suggested that a fully effective study in heart failure may require treatment to be optimized to maximum levels and that a period of 3 months’ treatment may then be required to eliminate any placebo effect and to demonstrate the full value of treatment. In the present study, the majority of patients were on the lowest effective dose of trial drug, and although the study was of 12 weeks’ duration a portion of this time was taken up by a dose-titration phase. If Packer’slo suggestions are correct, this study may have been too short and may have used doses too low to separate the effects of lisinopril and captopril completely satisfactorily. It is noteworthy that, with regard to overall NYHA grading, in the present study the 2 treatments were indistinguishable, both achieving an improvement in status of r35%, suggesting that on a global rating, both drug treatments were equivalent. Neither ACE inhibitor had any significant impact on the hourly rate of either ventricular ectopic counts or couplets, nor was there any difference between treatments with regard to the proportion of patients in which ventricular ectopic count rates were reduced. Despite the effect of both ACE inhibitor treatments on ventricular ectopic counts it is unlikely to be of clinical importance. In the earlier Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS),li enalapril treatment reduced mortality principally by reducing the frequency of deaths due to worsening heart failure rather than those attributed to sudden cardiac death, Considering the observations of the present study, and taken together with the finding of the CONSENSUS study, l1 the data suggest that, although ACE inhibitors will reduce the frequency of ventricular ectopic counts in a proportion of paOCTOBER
8, 1992
tients, this effect may be attributable primarily to their effects on the underlying pathologic condition causing the cardiac heart failure, rather than to a direct antiarrhythmic effect. Both once-daily lisinopril and twice-daily captopril were equally well tolerated and any adverse events observed were in line with those previously observed in heart failure patients treated with ACE inhibitors. Cough was infrequently reported in both treatment groups. An adverse effect on renal function, reflecting the effect of treatment on systemic and renal hemodynamics, may occur during ACE inhibitor treatment in some patients with CHF.12 In this regard, long-acting ACE inhibitors have been more frequently associated with increases in blood urea nitrogen and creatinine than have shorter-acting ACE inhibitors.3 However, in the present study the values for plasma potassium, urea, and creatinine remained fairly steady throughout for both lisinopril and captopril treatments; indeed the use of lisinopril was not associated with an increased frequency of elevations in potassium, urea, or creatinine compared with captopril. Taken together, these observations indicate that administration of once-daily lisinopril was not associated with an increased risk of renal impairment compared with the shorter-acting captopril administered twice daily. It is concluded that once-daily lisinopril5-20 mg is as effective as twice-daily captopril12.5-50 mg in terms of improvement of exercise capacity and changes in NYHA grading. Moreover, both drugs reduce the frequency of ventricular premature beats to an equal extent and are associated with a similar spectrum of adverse events in the heart failure patient. APPENDIX Members of the study group were as follows. Germany: Prof. A. Horsch (Speyer); Prof. D. Hey, Dr. K. Lindner (Bietigheim); Dr. P. Harbath, Dr. H.G. Gieretz (Bottrop); Dr. D. Storp, Dr. J. Heuse (Laatzen); Prof. P. Schuster (Eschweiler); Prof. J. Senges, Dr. M. Zander (Ludwigshafen); Dr. G. Pfeifer (Karlsruhe); Prof. J. Wagner, Dr. M. Pieper (Berlin); Prof. H.W. Hoepp, Dr. G. Mager, Dr. M. Hirsch (Koln); Prof. H. Simon, Dr. T. Ezzeldin (Dtiren); Prof. S. Gorlitz, Dr. R. Jacksch, Dr. A. Heidenblut (Essen); Prof. W. Kutscha, Dr. S. Roelcke (Landau); Dr. H.C. Hidajat (Rothenburg-
Fulda); Dr. A. Kemmling (Volklingen); Dr. W. Mansury (Neuwied); Dr. L. Szappanos (Darmstadt); Dr. J. Heck, Dr. H.-P. Hobbach (Altenkirchen). Italy: Prof. A. Mancini, Dr. A. Ferrara, Prof. N. Mininni, Dr. G. De Simone, Dr. S. Severino, Prof. M. Cafiero, Dr. F. Benvenuto, Prof. 0. De Divitiis, Dr. S. Di Somma (Naples); Prof. E. Correale, Dr. A. Brienza (Caserta); Prof. B. Ravera, Dr. A. Manaca (Salerno); Prof. L. Colonna, Dr. E. Fino (Bari); Prof. E. Geraci, Dr. F.P. Sabella (Palermo); Prof. R. Della Monica, Dr. A. Maliconico (Cara di Tirreui); Prof. F. Furlanello, Dr. G. Inama (Trento); Prof. G. Pinelli, Dr. G. Di Pasquale, Dr. A.M. Lusa, Dr. P. Grazi (Bologna); Prof. F. Rusticali, Dr. A. Catapano, Prof. C. Longhini, Dr. M. Vaccari, Dr. E. Baracca (Ferrara); Dr. L. Rossi, Dr. E. Carbonieri (Verona).
REFERENCES 1. Braunwald E. ACE inhibitors-a cornerstone of the treatment of heart failure. N Ellgl J&d 1991;325:351-353. 2. Cohn JN, Johnson G, Ziesche S, Cobb F, Francis G, Tristani F, Smith R, Dunkman WB, Loeb H, Wong M. A comparison of enalapril with hydtalazineisosorbide diiitrate in the treatment of chronic congestive heart failure. N Engl JMed 1991;325:303-310. 3. Packer M, Lee WH, Yashak M, Media N. Comparison of captopril and enalapril in patients with severe chronic heart failure. N Engl .I Med 1981;315: 847-853. 4. Giles TD, Katz R, Sullivan JM, Wolfson P, Haugland
M, KirIii P, Powers E, Rich S, Hackshaw B, Chiaramida A, Rouleau JL, Fisher MB, Pigeon J, Rush JE. Short- and long-acting angiotensin-converting enzyme inhibitors: A randomised trial of lisinopril versus captopril in the treatment of congestive heart failure. JAm Co[l Cardiol1989;13:1240-1247. 5. Harper RW, Cl&on H, Middlebrook K, Anderson S, Pitt A. The acute and chronic haemodynamic effects of prazosin in severe congestive heart failure. Me-d JAust 1980;2:36-38. 6. Kramer BL, Massie BM, Topic N. Controlled trial of captoptil in chronic heart failure. A rest and exercise haemodynamic study. Circulation 1983; 67(suppl4):807-816. 7. Markham RV, Corbett JR, Gihnore A, Pettinger WA, Fiih BG. Efficacy of prazosin in the management of chronic congestive heart failure: a 6-month randomised, double-blind, placebocontrolled study. Am J Cardbl1983;51:13461352. 8. Chalmers JP, West MJ, Cyran J, De la Torre D, EngIart M, Kramer M, Lewis GR, Maranhas MF, Myburgh DP, Shuster P. Placebo-controIIed study of lisinopril in congestive heart failure: a multicentre study. J Cardimmc Pharmacol1987;9(suppl3):S89S97. 9.ZwehI W, Rucinska El. Long term effects of lisinopril in patients with chronic heart failure: a multicentre, placebo controlled trial. In: Nicholls MG, ed. A Focus on the Clinical Effects of a Long Acting ACE Inhibitor/Heart Failure. New York: Raven Press, 1990:31400. 10. Packer M. Pathophysiological mechanisms underlying the effects of betaadrenergic agonists and antagonists on functional capacity and survival in chronic heart failure. Ctiulation 1990,82(suppl 1):1-77-I-88. il. CONSENSUS Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North Scandinavian EnaIapriI Survival Study (CONSENSUS). N Engl JMed 1987;316:1429-1435. 12. Packer M, Lee WH, Medma N, Yushak M, Kessler PD. Functional renal insufhciency during long-term therapy with captopril and enalaptil in severe chronic heart failure./lrvl Intern Med 1987;106:346-354.
A SYMPOSIUM:
HEART
FAILURE
MANAGEMENT
77c
MD,
Once-daily liiinopril(6-20 mg) was compared with twice-daily captopril(l2.6-60 mg) in a double-blind, randomized, parallel-group study of angiotensin-converting enzyme (ACE) inhibition conducted in 31 centers for 12 weeks in patients with heart failure (New York Heart Association class IHIl) who were currently receiving digitalis and / or diuretics. The drugs were compared with regard to their effects on exercise duration, measured with bicycle ergometry, and on ectopic actii, measured u&g Holter monitoring. Both drugs significantly increased exercise duration after both 6 and 12 weeks of randomized treatment. Neither ACE inhibitor had any significant impact on the hourly rate of either ventricular ectopic counts or couplets, nor was there any difference between treatments with regard to the proportions of patients in whom ventrkular ectopic counts were reduced. Both drugs were well tolerated, with no differences observed between treatments. Potassium, urea, and creatinine levels remained stable for both treatments throughout the study. (Am J Cardiol1662;70:7OC%77C)
From the Medizinische UnversitBtsklinik, Innere Medizin III, HomburgiSaar, Germany (R.B), and Institute of Cardiology, University of Verona, Verona, Italy (P.Z.). Address for reprints: Roland Bach, MD, Medizinische Unversitltsklinik, Innere Medizin III, 6650 HomburgiSaar, Germany.
70c
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
and Piero Zardini,
F
MD
ollowing the recent publication of several major studies on the use of inhibitors of angiotensin-converting enzyme (ACE), the use of these drugs in the treatment of moderate and severe congestive heart failure (CHF; New York Heart Association [NYHA] classes III and IV) has now become standard practice as an addition to the use of diuretic drugs and/or digitalis.l This relatively new class of agents is beginning to replace the use of other vasodilators, especially as they have been shown to have superior efficacy over some of the older vasodilator therapies.2 Despite the wide acceptance of ACE inhibitors in the treatment of CHF, there remains some debate as to the relative merits of short-acting and longacting drugs of this class.3>4 Packer et al3 have demonstrated that under certain conditions, relative hypotension and renal impairment were more frequently found in patients treated with a longacting ACE inhibitor, enalapril, than in those patients treated with a short-acting ACE inhibitor, captopril. Conversely, Giles et al4 demonstrated that once-daily lisinopril is at least as effective as 3 times daily captopril when these drugs are coadministered with digitalis and/or diuretics in patients with NYHA classes II-IV heart failure. Additionally, this study demonstrated that the administration of a long-acting ACE inhibitor once daily was not associated with a significantly greater risk of adverse events. The primary aim of the study reported here was to compare the efficacy of once-daily lisinopril with twice-daily captopril in the treatment of patients with mild-to-moderate heart failure (NYHA classes II and III), maintained on optimal doses of digitalis and/or diuretics. In addition, an investigation of the effect of these ACE inhibitors on ventricular ectopic activity, as measured by 24-hour electrocardiographic Holter monitoring, was undertaken. OCTOBER
8, 1992
METHODS Patients Male or female patients > 21 years old with the clinical signs and symptoms of CHF (NYHA classes II and III), were considered eligible for entry into the study. The diagnosis of CHF was confirmed by the use of chest radiograph (cardiothoracic ratio > OS), radioisotope scan, or echocardiography (ejection fraction < 45%). All patients had to be capable of performing an exercise protocol for 4-12 minutes on a bicycle ergometer. All were symptomatic on stable doses of either digitalis or diuretics or both. Patients were excluded from the study if they had a recent history of myocardial infarction or cardiac surgery (including percutaneous transluminal coronary angioplasty), cerebrovascular accident, clinically important renal disorders, or right heart failure. Patients with intrinsic lung disease limiting exercise performance were also excluded, as were those with arrhythmias that required therapy, other than with digoxin or amiodarone. Women capable of childbearing and patients with fixed rate cardiac pacemakers were also excluded. Patients with a history of liver, blood, or endocrine disorders or clinically significant abnormal levels of serum potassium were not permitted to enter the study, as were those with any known hypersensitivities or other contraindications to ACE inhibitors. Patients with a known history of drug or alcohol abuse were also excluded, and in some centers patients who had reduced absorption following gastrointestinal surgery or those who had taken part recently in another medical study were also excluded. Patients were informed of the aims, methods, possible benefits, and hazards of the study, and gave their informed consent to enter into the study. This study was conducted in accordance with the Declaration of Helsinki (revised Venice 1983) and local ethics committee approval was obtained. Study design: The study was a double-blind, parallel-group, randomized, multicenter design. The study incorporated a placebo baseline period of lo-14 days, during which time the digoxin and/or diuretic doses were optimized according to the patients’ symptoms, and all other vasodilator and ACE inhibitor therapies were withdrawn. Patients were then randomized to either lisinopril or captopril for 12 weeks. The randomization was carried out using a separate randomization scheme, which was generated for each participating center. At the end of the baseline period, patients were assigned to a treatment group, using the computergenerated random scheme for that center. During the run-in period, placebo tablets (matching 2.5 mg
lisinopril taken once daily, and matching 12.5 mg captopril taken twice daily) were administered to the patients single blind. The initial dose of the double-blind treatment was given as a single oral dose of either 2.5 mg lisinopril or 6.25 mg captopril. Any symptoms reported or experienced by the patient were fully documented. To achieve double-blind treatment during the active treatment period, a double-dummy technique was employed. On each day patients who received lisinopril also took the appropriate number of placebo tablets matching captopril tablets and vice versa. In the case of patients randomized to lisinopril, after monitoring the effect of the initial dose (2.5 mg), 1 lisinopril tablet (5 mg) was taken per day. If, after 2 weeks of this treatment, there was evidence of clinical improvement, then the patient continued on that dosage thereafter. An increase in the dose of study medication was considered after 2 weeks of randomized treatment, and again after a further 2 weeks of randomized treatment if the following 2 safety conditions were met: (1) standing systolic blood pressure was 2 90 mm Hg, and (2) there were no symptoms of hypotension (syncope, faintness, orthostatic dizziness) on the current dose and there was evidence of a need for additional therapeutic effect in the opinion of the investigators. If the dosage of study medication needed to be increased after 2 weeks, lisinopril dosage was increased to 10 mg once daily and, again, after a further 2 weeks of treatment lisinopril could be increased to a maximum of 20 mg once daily. The dose of lisinopril was reduced at any relevant visit if (in the view of the investigator) the patient was exhibiting symptomatic hypotension or any other adverse event. For patients who received captopril, 1 captopril tablet (12.5 mg) was taken twice daily after monitoring the effect of the initial dose (6.25 mg). If after 2 weeks’ randomized treatment there was evidence of clinical improvement, then the patient continued on a dosage of captopril 12.5 mg twice daily thereafter. An increase in the dose of the study medication to 25 mg twice daily was considered after 2 weeks of randomized treatment if the previous safety conditions and the need for further efficacy were met. If after a further 2 weeks of treatment the dosage of study medication needed to be increased, captopril doses were increased to a maximum of 50 mg twice daily. As was the case for the lisinopril-treated group of patients, the dosage of captopril could also be reduced at the discretion of the investigator if the symptoms so indicated. A SYMPOSIUM:
HEART FAILURE
MANAGEMENT
71c
Concurrent medication: Patients taking digitalis were to be maintained on a constant dose throughout the study. If evidence of toxicity appeared, the dose was either reduced or treatment discontinued. The morning dose of diuretic was withheld on the day the patient received the initial dose of randomized treatment. Thereafter, the diuretic dose was adjusted during the randomizedtreatment phase as needed, based on the clinical assessment. Potassium-sparing diuretics were prohibited during the study. Serum potassium was anticipated to remain unchanged on ACE inhibitor treatment, or to increase slightly during the course of the study. However, if a patient developed hypokalemia, potassium supplements were administered. Additionally, sublingual nitroglycerin was taken as required. Chronic nitrate therapy, when used, was kept constant throughout the study. Concurrent therapy with anticoagulants was permitted; no other drugs known to affect preload or afterload could be administered during the course of this study. Clinical assessment: A thorough physical examination of each patient was conducted at the time of entry into the study. An abbreviated symptom review and physical examination was performed at randomization, and this was repeated at the end of 12 weeks of randomized treatment. Exercise testing was performed using an electrically braked bicycle ergometer and followed a schedule with continuous incremental workloads. All patients underwent a bicycle exercise test on 2 separate occasions during the placebo run-in, prior to entry into the randomized treatment phase of the study. The first exercise test served as a familiarization procedure for the patient, the second exercise test enabled the duration of exercise to be determined. Exercise tests were also carried out after 6 and 12 weeks of treatment. Patients undertook each exercise test at the same time of day (preferably in the morning), and medication was delayed on the day of the exercise test until measurements were complete. Blood pressure was measured at each clinical visit on the same arm using a mercury sphygmomanometer. Heart rate was measured on each visit by palpation of the radial pulse over a 30-second period. Recordings of blood pressure and heart rate were made after the patient had been standing for 2 minutes. Classification of the patient according to the NYHA criteria was done by the same observer throughout the study and it was assessed at the end of the placebo run-in and after 12 weeks of treatment. 72C
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
Holter monitoring was performed for a 24-hour period during placebo run-in and during the last week of treatment. The 24-hour electrocardiographic recordings were made using a 2-channel Holter monitor. Analysis of all 24-hour tapes was performed electronically by Hertford Medical (Hertford, UK). Blood samples were obtained on entry to the study and after 12 weeks of treatment for hematology and biochemistry assessments, and at weeks 2, 4, and 6 of randomized treatment, for biochemistry assessment only. Full blood count and assessment of plasma urea, sodium, potassium, and creatinine were carried out. Volunteered adverse events were obtained at each visit after the start of the placebo run-in period. In response to the question, “Have you had any symptoms or problems since your last visit?” the nature of any symptoms were recorded. Patients were asked to bring any unused tablets with them at each visit to assess compliance. Statistical analysis: Statistical analysis was directed at estimating differences between the effects of lisinopril and captopril on exercise capacity after 12 weeks of therapy. Analysis of covariante was used to compare lisinopril and captopril for the change in exercise duration from the run-in visit to treatment week 6 and the change from the run-in visit to treatment week 12. The initial analysis of covariance took account of the following sources of variation: center, covariate, treatment, and center-by-treatment interaction. The exercise duration at the run-in visit was used as the covariate. The influence of any patients withdrawing from the study or violating the protocol was investigated and the data used in the most appropriate manner to minimize bias. Previously published studies give an estimate of the variability of exercise duration data from bicycle exercise tests5-7 and indicate a between-patient standard deviation from 1.4-3.1 minutes. To detect a minimum difference of 60 seconds in exercise duration between the 2 treatment groups, it was calculated that 172 patients (86 per group) would be required. This calculation assumed a 5% level of significance with 90% power, and assumed a standard deviation of 120 seconds. To allow for withdrawals it was planned that 250 patients (125 per group) were required for this study. Three analyses of exercise duration were carried out: completers analysis, per protocol analysis, and last value analysis. Of these analyses, the completers analysis assessed the treatment effects, ignoring potential bias or loss of precision
OCTOBER
8, 1992
due to withdrawals (13% of patients in this study) and protocol violators and deviators. In this regard, it is close to an analysis based on the “intention to treat” principle. The per protocol analysis fulfilled the study objectives with respect to the protocol. The East value analysis attempted to take account of the withdrawals in estimating the treatment effects. In this article, the results of the completers analysis are reported because the results and conclusions were the same as those obtained from the per protocol
analysis.
Data on arrhythmias tend to have high variability, and the extent to which any treatment may reduce the incidence of arrhythmias is uncertain as is the clinical importance of any given reduction in CHF. Before any analysis could be undertaken, it was necessary to adjust the raw data recorded, because the individual periods of electrocardiographic monitoring were not exactly 24 hours and in many cases were much less than 24 hours. To allow for variations in the duration of electrocardiogram recording between different centers, the electrocardiographic data were converted to hourly rate for each recording, by dividing the values observed for the entire monitoring period by the duration of monitoring (hours). Only data from patients with a minimum of 8 hours of recording for both run-in and end of treatment assessment were summarized. Data from patients failing to satisfy this criterion were excluded from analyses of the electrocardiogram. In this report the results of the analysis based on patients completing the study alone are described, because the results and conclusions are the same as those obtained from a per protocol analysis. The skewed nature of the ventricular ectopic counts, couplet, and nonsustained ventricular tachycardia data suggested that it would be inappropriate to perform statistical analyses using parametric analysis of variance or covariance. These data were, therefore, analyzed using the nonparametric Wilcoxon rank-sum test. The proportion of patients in each treatment group with a reduction in the hourly rate of ventricular ectopic counts from run-in to week 12 was compared using Fisher’s exact test. The treatment effect was also assessed by summarizing the proportion of patients in each treatment group at week 12 whose ventricular ectopic activity was reduced. Two analyses of ventricular ectopic count, couplets and nonsustained ventricular tachycardia were carried out: completers analysis andperprotocol analysis, based on the principles outlined above. All safety data collected during the course of the
TABLE I Demographic Variables to Lisinopril or Captopril Variable Number patients Males (%I Females (%)
for the Patients Lisinopril
Captopril
148 117 (79) 31 (21)
108 (78) 31 (22)
Mean age (yr)
59
Age range (yr)
29-83
Body weight
Randomized
139
59
33-82 74.5
77.2
(kg)
Etiology of heart failure* (%I lschemic heart disease Cardiomyopathy Valvular heart disease Hypertension Other
52 35
49 41 8 18 5
14
22 9
*Some patients had more than 1 contributqfactorfor
heartfallure.
study were collated and the proportion of patients who reported adverse events and those who withdrew due to adverse events, were compared for both lisinopril and captopril treatments, using Fisher’s exact test. Because these analyses were concerned with treatment safety, data from all randomized patients were included. RESULTS A total of 315 patients entered the study; 28 of these withdrew during the placebo run-in period and 38 withdrew during the randomized treatment period. The demographic details of the patients randomized to treatment with lisinopril or captopril are given in Table I, which shows that both groups of patients were well matched with respect to demographic variables. At the end of the study, 43% of captopril patients were receiving 25 mg daily, 34% received 50 mg daily, and 23% were taking 100 mg daily. In the lisinopril-treated group the 5, 10, and 20 mg daily doses were being taken by 44,38, and 18% of the patients, respectively. The effects of lisinopril and captopril treatment on exercise duration for weeks 6 and 12 of randomized treatment are shown in Figure 1. At 6 weeks, the exercise duration was increased with both treatments. The increase was slightly greater for lisinopril than for captopril. However, this relatively small difference (3 seconds) was not statistically significant. Exercise duration increased by 47.2 seconds on lisinopril and 44.3 seconds on captopril (p = 0.77). Similarly, after 12 weeks of randomized treatment, exercise duration was increased by both lisinopril and by captopril. The increase seen at week 12 was slightly greater for lisinopril than for captopril, but the difference (5 seconds) was not statistically significant (p = 0.68). When compared with initial run-in values on placebo, both lisinopril and captopril treatment inA SYMPOSIUM:
HEART
FAILURE
MANAGEMENT
73c
* - * Captopril O---+ Lisinopril 600
T
_I1
(117) (117)
(115)
__---
-xc
(117) (117)
(116;
c c c r *
I
1.
I Run-in
1
I 6 weeks
I 12 weeks
FIGURE 1. The effect of liiinopril and captopril on exercise capacity at the end of the run-in period and after 6 and l2 weeks of treatment. Numbers in parentheses indicate the number of patients In each group.
creased exercise duration at both 6 and 12 weeks significantly (p = 0.0001) in all cases. 24.hour electrocardiographic monitoring: There was no statistically significant difference between the treatments for the rates of ventricular ectopic counts or couplets. The median hourly rate for ventricular tachycardia was 0. Of 90 patients who received lisinopril, 51 (57%) and 50 of 93 patients (54%) who received captopril demonstrated a reduction in the hourly rate of ventricular ectopic counts from placebo run-in to the week 12 assessment. No significant difference between the 2 treatment groups was detected (p = 0.77). Symptom review and physical examination: Figure 2 shows the percentage of randomized patients presenting with a particular symptom of 50
40
1
heart failure both at placebo run-in assessment and after 12 weeks of randomized treatment. There was an overall improvement in the monitored symptoms following treatment for both groups of patients. With regard to the percentage of patients improving, the effect of treatment was similar for both captopril and lisinopril. Mean standing blood pressures and heart rate recorded at each visit remained consistent throughout the study in both treatment groups. No effects of treatment on the mean blood pressures or heart rate were observed. Evaluation of NYHA class: Of the 125 lisinopril patients completing the study, 44 (35%) showed an improvement in NYHA status, 79 (63%) presented no change, and only 2 (1.6%) showed a deterioration in their condition. Of the 127 captopril patients completing the study, 51 (40%) showed improvement in their condition compared with 74 (58%) who showed no change and 2 (1.6%) who deteriorated (Figure 3). Adverse events: Adverse events were reported for 23 of 148 patients (16%) who received lisinopril, and 21 of 139 patients (15%) who received captopril. No significant difference in the incidence of adverse events was seen between the 2 active treatment groups. Nine of 148 patients (6%) who were randomized to lisinopril and 7 of 139 patients (5%) who were randomized to captopril withdrew from the study due to adverse events (including death). The reasons for withdrawal are shown in Table II. There were no significant differences detected between the captopril and lisinopril treatment groups. Five patients treated with lisinopril
Lisinopril
run-in
12 weeks
Lisinopril
0
Captopril
run-in
12 weeks
Captopril
m
FIGURE 2. Percentage of patients with symptoms of heart failure present at run-in and after I2 weeks of treatment with either lisinopril or captopril.
74c
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
OCTOBER
8. 1992
and 2 patients treated with captopril died during the randomized phase of the trial. The causes of death in lisinopril-treated patients were pulmonary edema, ventricular fibrillation, sudden death (n = 2), and an accident. The 2 deaths on captopril treatment were due to cardiac failure following acute myocardial infarction and pulmonary edema. None of these deaths was considered to be related to treatment with the ACE inhibitors. Three patients died during the prerandomization phase of this study due to ventricular fibrillation (n = 1) and sudden death (n = 2). A total of 15 of 148 patients (10%) who received lisinopril and 15 of 139 patients (11%) who received captopril experienced adverse events that did not lead to withdrawal from the study. No obvious treatment-related effects were observed in the laboratory parameters for either the lisinoprilor captopril-treated groups. Further, individual patient data showed no apparent treatment-related effects from either of these 2 drugs. Two patients who were treated with lisinopril showed changes in neutrophil count. For 1 of these patients, this value was above the laboratory normal range at entry and fell below the laboratory normal range after 12 weeks of treatment with lisinopril. The second patient displayed a decrease in neutrophil count but the value remained within the normal range throughout the study. Mean plasma potassium concentration was unchanged after 12 weeks of treatment with captopril, 4.33 & 0.44 mEq/liter at entry, compared with 4.35 -+ 0.49 mEq/liter after 12 weeks. In contrast, the mean plasma potassium concentration increased on lisinopril treatment from 4.30 c 0.43 mEq/liter at entry to 4.38 & 0.39 mEq/liter at the
TABLE II Withdrawal
Adverse Events Including of Treatment*
Death Leading
Lisinopril Deaths
Other
Pulmonary Accident
Captopril
edema
Pulmonary edema Cardiac failure due to acute myocardial infarction
Ventricular fibrillation Sudden death (2 patients) Abdominal pain Tingling sensation Headaches Myocardial infarction Coughing Dyspnea Blood urea nitrogen increased Hyperkalemia
events
to
Prurigo Exanthema Vertigo Feeling unwell Headaches Taste unusual Coughing Dizziness
*Some patients had more than 1 adverse event leading to withdrawal.
end of 12 weeks of treatment. Per patient, 39% of captopril-treated patients and 36% of lisinopriltreated patients showed a decrease in plasma potassium of > 0.5 mEq/liter and 53% of captopril and 56% of lisinopril patients showed an increase of plasma potassium >0.5 mEq/liter. These changes, however, were not significant, and there were no differences between treatments. Plasma urea concentrations fell slightly, but not significantly, after lisinopril and captopril treatment. At entry the mean plasma urea concentrations were 6.42 + 2.68 and 6.13 + 2.69 mmol/liter for lisinopril and captopril, respectively. After 12 weeks of treatment the plasma urea concentrations were 6.32 +- 2.48 mmol/liter for lisinopril and 5.92 5 2.35 mmol/liter for captopril. Of the captopril-treated patients, 11% showed a decrease in plasma urea concentrations greater than the predefined limit of 1.7 mmol/liter and an increase in
r
FIGURE 3. Change in functional status of patients completing l2 weeks of treatment, as assessed by New York Heart Association c[radW
Captopril
Lisinopril w
improved
c>
worsened
no change
A SYMPOSIUM:
HEART
FAILURE
MANAGEMENT
7!%
plasma urea concentration greater than 1.7 mmol/ liter was seen in 13%. The corresponding figures for the lisinopril-treated group were 13% and 14%, respectively. The mean plasma creatinine concentration was not significantly increased from 92.9 2 19.2 umol/ liter at entry to 94.0 + 20.4 pmol/liter after 12 weeks of treatment with lisinopril. Mean plasma creatinine concentration fell on captopril treatment from 93.1 & 20.2 pmol/liter at entry to 93.0 + 19.3 u,mol/liter during 12 weeks of treatment. A decrease in plasma creatinine concentration greater than the predefined limit of 17.7 pmol/liter was seen in 10% of lisinopril- and 9% of captopril-treated patients and increases in plasma creatinine >26.5 u,mol/liter were seen in 8% of both captopril- and lisinopril-treated patients. Other routine biochemical analyses indicated no obvious treatment-related effects for either lisinopril- or captopril-treated groups. The individual patient biochemical data did not show any obvious treatment-related effects. No discernible differences between the treatment groups were seen. DISCUSSION This study demonstrates the equivalence of once-daily lisinopril and twice-daily captopril in patients with mild to moderate heart failure. Both treatments produced significant, but very similar, improvements in the exercise capacity of the patients. A previous study in which once-daily lisinopril was compared with 3 times daily captopril showed a greater improvement in exercise capacity in lisinopril-treated patients compared with the captopril-treated patients (using a per protocol an~lysis).~ The results of this study appear to be at variance with this earlier work. However, there may be several reasons for the apparent lack of difference between the 2 treatments seen in this present study. This study employed bicycle ergometry to assess exercise capacity rather than a treadmill test. A previous study in which lisinopril was compared with placebo, and which used both treadmill and bicycle ergometry testing, produced different exercise times for these 2 methods with smaller differences between the ACE inhibitor and placebo for the bicycle ergometry.* However, larger differences have been observed in other placebocontrolled studies of lisinopril.9 The shorter exercise times observed using bicycle ergometry may make any small differences between treatments more difficult to detect than by using the longer exercise times of the treadmill tests. 76C
THE AMERICAN
JOURNAL
OF CARDIOLOGY
VOLUME
70
In the present study the time of assessment of exercise capacity in relation to the timing of the previous dose of medication was different from that reported in a previous comparative study.4 In the latter study, exercise capacity was measured during the day at no particularly fixed relation to the timing of the last dose of the study medication. However, in the present study the exercise capacity was assessed at least 24 hours after the last lisinopril dose and at least 12 hours after the last captopril dose. Under the conditions of this study, it is anticipated that the blood (and tissue) levels of both drugs would be at their respective nadirs at the time of exercise assessment, perhaps minimizing any differences between the treatments that might have been observed had blood concentrations of the respective drugs been higher. Packer,lO has suggested that a fully effective study in heart failure may require treatment to be optimized to maximum levels and that a period of 3 months’ treatment may then be required to eliminate any placebo effect and to demonstrate the full value of treatment. In the present study, the majority of patients were on the lowest effective dose of trial drug, and although the study was of 12 weeks’ duration a portion of this time was taken up by a dose-titration phase. If Packer’slo suggestions are correct, this study may have been too short and may have used doses too low to separate the effects of lisinopril and captopril completely satisfactorily. It is noteworthy that, with regard to overall NYHA grading, in the present study the 2 treatments were indistinguishable, both achieving an improvement in status of r35%, suggesting that on a global rating, both drug treatments were equivalent. Neither ACE inhibitor had any significant impact on the hourly rate of either ventricular ectopic counts or couplets, nor was there any difference between treatments with regard to the proportion of patients in which ventricular ectopic count rates were reduced. Despite the effect of both ACE inhibitor treatments on ventricular ectopic counts it is unlikely to be of clinical importance. In the earlier Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS),li enalapril treatment reduced mortality principally by reducing the frequency of deaths due to worsening heart failure rather than those attributed to sudden cardiac death, Considering the observations of the present study, and taken together with the finding of the CONSENSUS study, l1 the data suggest that, although ACE inhibitors will reduce the frequency of ventricular ectopic counts in a proportion of paOCTOBER
8, 1992
tients, this effect may be attributable primarily to their effects on the underlying pathologic condition causing the cardiac heart failure, rather than to a direct antiarrhythmic effect. Both once-daily lisinopril and twice-daily captopril were equally well tolerated and any adverse events observed were in line with those previously observed in heart failure patients treated with ACE inhibitors. Cough was infrequently reported in both treatment groups. An adverse effect on renal function, reflecting the effect of treatment on systemic and renal hemodynamics, may occur during ACE inhibitor treatment in some patients with CHF.12 In this regard, long-acting ACE inhibitors have been more frequently associated with increases in blood urea nitrogen and creatinine than have shorter-acting ACE inhibitors.3 However, in the present study the values for plasma potassium, urea, and creatinine remained fairly steady throughout for both lisinopril and captopril treatments; indeed the use of lisinopril was not associated with an increased frequency of elevations in potassium, urea, or creatinine compared with captopril. Taken together, these observations indicate that administration of once-daily lisinopril was not associated with an increased risk of renal impairment compared with the shorter-acting captopril administered twice daily. It is concluded that once-daily lisinopril5-20 mg is as effective as twice-daily captopril12.5-50 mg in terms of improvement of exercise capacity and changes in NYHA grading. Moreover, both drugs reduce the frequency of ventricular premature beats to an equal extent and are associated with a similar spectrum of adverse events in the heart failure patient. APPENDIX Members of the study group were as follows. Germany: Prof. A. Horsch (Speyer); Prof. D. Hey, Dr. K. Lindner (Bietigheim); Dr. P. Harbath, Dr. H.G. Gieretz (Bottrop); Dr. D. Storp, Dr. J. Heuse (Laatzen); Prof. P. Schuster (Eschweiler); Prof. J. Senges, Dr. M. Zander (Ludwigshafen); Dr. G. Pfeifer (Karlsruhe); Prof. J. Wagner, Dr. M. Pieper (Berlin); Prof. H.W. Hoepp, Dr. G. Mager, Dr. M. Hirsch (Koln); Prof. H. Simon, Dr. T. Ezzeldin (Dtiren); Prof. S. Gorlitz, Dr. R. Jacksch, Dr. A. Heidenblut (Essen); Prof. W. Kutscha, Dr. S. Roelcke (Landau); Dr. H.C. Hidajat (Rothenburg-
Fulda); Dr. A. Kemmling (Volklingen); Dr. W. Mansury (Neuwied); Dr. L. Szappanos (Darmstadt); Dr. J. Heck, Dr. H.-P. Hobbach (Altenkirchen). Italy: Prof. A. Mancini, Dr. A. Ferrara, Prof. N. Mininni, Dr. G. De Simone, Dr. S. Severino, Prof. M. Cafiero, Dr. F. Benvenuto, Prof. 0. De Divitiis, Dr. S. Di Somma (Naples); Prof. E. Correale, Dr. A. Brienza (Caserta); Prof. B. Ravera, Dr. A. Manaca (Salerno); Prof. L. Colonna, Dr. E. Fino (Bari); Prof. E. Geraci, Dr. F.P. Sabella (Palermo); Prof. R. Della Monica, Dr. A. Maliconico (Cara di Tirreui); Prof. F. Furlanello, Dr. G. Inama (Trento); Prof. G. Pinelli, Dr. G. Di Pasquale, Dr. A.M. Lusa, Dr. P. Grazi (Bologna); Prof. F. Rusticali, Dr. A. Catapano, Prof. C. Longhini, Dr. M. Vaccari, Dr. E. Baracca (Ferrara); Dr. L. Rossi, Dr. E. Carbonieri (Verona).
REFERENCES 1. Braunwald E. ACE inhibitors-a cornerstone of the treatment of heart failure. N Ellgl J&d 1991;325:351-353. 2. Cohn JN, Johnson G, Ziesche S, Cobb F, Francis G, Tristani F, Smith R, Dunkman WB, Loeb H, Wong M. A comparison of enalapril with hydtalazineisosorbide diiitrate in the treatment of chronic congestive heart failure. N Engl JMed 1991;325:303-310. 3. Packer M, Lee WH, Yashak M, Media N. Comparison of captopril and enalapril in patients with severe chronic heart failure. N Engl .I Med 1981;315: 847-853. 4. Giles TD, Katz R, Sullivan JM, Wolfson P, Haugland
M, KirIii P, Powers E, Rich S, Hackshaw B, Chiaramida A, Rouleau JL, Fisher MB, Pigeon J, Rush JE. Short- and long-acting angiotensin-converting enzyme inhibitors: A randomised trial of lisinopril versus captopril in the treatment of congestive heart failure. JAm Co[l Cardiol1989;13:1240-1247. 5. Harper RW, Cl&on H, Middlebrook K, Anderson S, Pitt A. The acute and chronic haemodynamic effects of prazosin in severe congestive heart failure. Me-d JAust 1980;2:36-38. 6. Kramer BL, Massie BM, Topic N. Controlled trial of captoptil in chronic heart failure. A rest and exercise haemodynamic study. Circulation 1983; 67(suppl4):807-816. 7. Markham RV, Corbett JR, Gihnore A, Pettinger WA, Fiih BG. Efficacy of prazosin in the management of chronic congestive heart failure: a 6-month randomised, double-blind, placebocontrolled study. Am J Cardbl1983;51:13461352. 8. Chalmers JP, West MJ, Cyran J, De la Torre D, EngIart M, Kramer M, Lewis GR, Maranhas MF, Myburgh DP, Shuster P. Placebo-controIIed study of lisinopril in congestive heart failure: a multicentre study. J Cardimmc Pharmacol1987;9(suppl3):S89S97. 9.ZwehI W, Rucinska El. Long term effects of lisinopril in patients with chronic heart failure: a multicentre, placebo controlled trial. In: Nicholls MG, ed. A Focus on the Clinical Effects of a Long Acting ACE Inhibitor/Heart Failure. New York: Raven Press, 1990:31400. 10. Packer M. Pathophysiological mechanisms underlying the effects of betaadrenergic agonists and antagonists on functional capacity and survival in chronic heart failure. Ctiulation 1990,82(suppl 1):1-77-I-88. il. CONSENSUS Trial Study Group. Effects of enalapril on mortality in severe congestive heart failure. Results of the Cooperative North Scandinavian EnaIapriI Survival Study (CONSENSUS). N Engl JMed 1987;316:1429-1435. 12. Packer M, Lee WH, Medma N, Yushak M, Kessler PD. Functional renal insufhciency during long-term therapy with captopril and enalaptil in severe chronic heart failure./lrvl Intern Med 1987;106:346-354.
A SYMPOSIUM:
HEART
FAILURE
MANAGEMENT
77c
