European Heart Journal (1991) 12, 705-711

Effects of quinapril on exercise tolerance in patients with mild to moderate heart failure G. A. J.

RIEGGER

Quinapril Research Group, Medizinische Universitiitsklinik, Wiirzburg , Germany KEY WORDS: Angiotensin converting enzyme inhibition, mild heart failure The objective of this double-blind, placebo-controlled, randomized multicentre study was to determine whether treatment with the new non-sulfhydryl angiotensin converting enzyme (ACE) inhibitor quinapril, as an addition to maintenance therapy with digitalis and/or diuretics, would improve exercise tolerance and patients' symptomatology over a treatment period of 3 months. Two hundred and twenty-five patients with mild to moderate heartfailure were studied infour parallel treatment groups receiving either placebo or quinapril in a dose of 10,20, or 40 mg day :', Compared to placebo, quinapril therapy resulted in a significant improvement in exercise time. The improvement in exercise tolerance was dose-related and showed a significant increase at the end ofthe study in the patients who completed the trial and in an intent-to-treat analysis. Twenty-six patients were on mono therapy with quinapril without concomitant medication. In this subgroup of patients, the increase in exercise time was comparable to the subgroup of patients on diuretic treatment alone. After the 12-week study, 189 patients entered an open-label trialfor 12 months, in which the dose of the ACE inhibitor could be adjusted according to clinical response. Despite a reduction in the daily dose ofquinapril, the patients reached the same level ofexercise capacity as in the 3-month study. No serious side effects were recorded, particularly no symptomatic hypotension or deterioration ofrenal function. The results of the study show that quinapril has a significant favourable effect on exercise tolerance and symptoms in patients with mild to moderate heart failure.

Introduction Vasodilator therapy for patients with congestive heart failure has become accepted over the past decade. Directacting vasodilators may produce acute improvement in the haemodynamic situation, but commonly fail to produce clinical benefits'!', Recently, the combination of isosorbide dinitrate and hydralazine has been shown to reduce mortality in patients with chronic congestive heart failure'", Several controlled and uncontrolled clinical trials[3-9] have demonstrated that captopril and enalapril induce short- and long-term haemodynamic improvement in patients with heart failure, and it has been suggested from pooled data of placebo-controlled trials'!" that converting enzyme inhibitors may beneficially influence the prognosis of patients with heart failure. Recently, it was shownl'!' that enalapril added to conventional therapy can reduce mortality and improve symptoms in patients with severe congestive heart failure, and captopril was found significantly more effective than placebo in improving exercise time and New York Heart Association class in patients with mild to moderate heart failure on maintenance therapy with diuretics. Converting enzyme inhibition may therefore be an alternative treatment to digoxin in this subset of patients'!", Submitted for publication on 10 November 1989,and in revised form 8 August 1990. Supported by Warner-Lambert Company, Ann Arbor, Michigan, U.S.A. Correspondence: Professor G. A. J. Riegger, MD, Medizinische Universitats-

klinik, Josef-Schneider-Stral3e 2,0-8700 Wiirzburg, Germany.

o195-668X/91/060705 + 07 $03.00/0

The primary objective of this randomized, parallel, double-blind, multicentre trial was to determine whether treatment with the new non-sulfhydryl angiotensin converting enzyme (ACE) inhibitor quinapril (WarnerLambert Company, Ann Arbor, Michigan/USA), which has shown beneficial acute haemodynamic effects[13] would improve exercise tolerance over a period of 3 months in patients with chronic mild to moderate heart failure. This question was addressed for the first time by evaluating three different doses of an ACE inhibitor in comparison with placebo when added to conventional therapy with digitalis and/or diuretics. After the double-blind study, the long-term efficacy of quinapril was investigated in a oneyear open label trial, where the ACE inhibitor dose could be adjusted according to clinical response.

Methods STUDY DESIGN

Three month study The efficacy and safety of orally administered quinapril 5 mg b.i.d., 10 mg b.i.d., and 20 mg b.i.d, were compared in a placebo-controlled, double-blind, parallel, randomized study in outpatients with congestive heart failure, New York Heart Association functional class II or III. F ollowing a 2-week placebo baseline phase, the patients who qualified for entry into the double-blind phase were randomly assigned to one of four parallel treatment groups for a treatment period of 12 weeks. Group 1 received placebo, group 2 quinapril 5 mg b.i.d., group 3 © 1991 The European Society of Cardiology

706

G. A. J. Riegger

Table 1

Patient characteristics (all patients)

Number (%) of patients Sex Men Women Median Age (years) Height (em) Duration ofCHF (years) Aetiology of CHF (%) CHD HHD Dilated cardiomyopathy Others

Placebo (n = 56)

Quinapril 5 mg b.i.d. (n = 57)

Quinapril 10 mg b.i.d. (n= 56)

Quinapril 20mgbj.d. (n = 56)

26 (46%) 30 (54%)

20 (35%) 37 (65%)

25 (45%) 31 (55%)

20 (36%) 36 (64%)

67 169

67 165

66 168

65 168

4

4

3

4

41 (73%) 34 (61 %) 6 (11 %) 3 (5%)

40 32 3 7

(70%) (56%) (5%) (12%)

35 35 6 11

(63%) (63%) (11%) (20%)

35 36 4 7

(63%) (64%) (7%) (13%)

CHF: congestive heart failure: CHD: coronary heart disease; HHD: hypertensive heart disease.

quinapril 10 mg b.i.d., and group 4 quinapril20 mg b.i.d. The quinapril dose was started at 5 mg b.i.d. and doubled at weekly intervals in groups 3 and 4 until the final dose was reached. The doses were administered at about 0800h and 2000h. Digitalis and non-potassium sparing diuretics were kept constant throughout the study. Clinical visits were scheduled weekly during the placebo baseline phase and at beginning 'of weeks 1, 2, 3, 4, 6, 8, 10, 12 and 13 of the double-blind phase. Since exercise tolerance was the major objective of the study, each patient performed four exercise tests, the first at selection, the second at the end of the placebo baseline phase, the third at the beginning of week 6, and the fourth at the beginning of week 13. All exercise tests were performed before the morning dose of medication, which meant about 12 h after the last dose. Exercise tolerance was evaluated by bicycle ergometry. Patients cycling in the supine position started with 25 W, patients cycling in the upright position with 50 W. The workload was increased in 25-W steps every 2 min. The patients exercised until they were unable to continue because of dyspnoea and fatigue. The position of exercising was kept constant throughout the study. Electrocardiogram, blood pressure and heart rate were obtained at the end of each workload. Classification according to the New York Heart Association was determined by the investigators at each clinical visit. A chest X-ray was performed at baseline and at the end of the study.

One-year study The controlled 3-month study was followed by a oneyear open label trial in which 189 patients participated. All patients started with a 5-mg dose of quinapril b.i.d. The dose could be increased up to 20 mg b.i.d. at 2-week intervals according to clinical response criteria. The dose of concomitant medication was held constant throughout the study period. The same exercise tests as used in the 3month study were performed after 3,6,9, and 12 months.

Patient selection Patients suffering from chronic congestive heart failure with symptoms that corresponded to New York Heart Association functional class II or III were selected. Patient characteristics and clinical status are summarized in Tables 1 and 2. Exercise time was to be at least 2 min limited only by dyspnoea and fatigue at the end of the placebo run-in phase. The cardiothoracic ratio had to be > 0·5. Patients whose ability to exercise was limited by induced arrhythmias or angina were excluded. Other criteria of exclusion were: patients with rapid deterioration of heart failure, myocardial infarction in the past 3 months, mitral stenosis, aortic stenosis, obstructive lesions of the left ventricular outflow tract, congestive heart failure due to non-cardiac aetiology (anaemia, thyrotoxicosis, etc.) and resting systolic blood pressure less than 100 mmHg. Concomitant therapy with vasodilators, beta-adrenergic blockers, calcium antagonists, long-acting nitrates, other ACE inhibitors, or potassiumsparing diuretics was prohibited. For patients taking medication likely to interfere with the trial there was a wash-out period of 1 week before the beginning of the placebo baseline phase.

Data analysis The primary variable for evaluation of efficacy was exercise time. The baseline value was measured at the end of the placebo run-in phase. Drug response was defined as an increase in exercise time of at least 20% from baseline. The changes from baseline at the endpoint were analysed inferentially by a covariance model (testing the effects of treatment, study centre, exercise time at baseline, digitalis, diuretics, age, sex, and interaction between treatment and study centre). If the effects of treatment groups were significantly different, the comparisons between all possible pairs of treatment groups were made using the Bonferroni-Holm procedure'!" Adjusted mean changes were presented.

Effects ofquinapril on ex ercise tolerance

Table 2

707

Clinical status at the start ofdouble-blind phase (patients who completed the study )

D yspnoea at rest (%) Dyspnoea on exertion (%) Fatigue(%) Paroxysmal nocturnal dy spnoea (%) Orthopnea (%) NYHA II (%J NYHA III ( %) Sinus rhythm (%) Atrial fibrillation (%) Cardiothoracic ratio (mean) Exercise time (s) (mean)

Quinapril 5mgb.i.d . (n =51)

Quinapril 10mgb.i.d. (n =48)

Quinapril 20mg b.i.d. (n=51)

15 98 74

12 98 76

19 100 75

98 76

32 9 53 47 82 18

24

25

10

10

53 47 86 14

58 42 88 12

27 18 67 33 93 7

0·58

0·54

0·56

0·51

245

247

251

267

Placebo (n = 4 7)

10

The Cochran-Mantel-Haenszel procedure'P ' was used for the inferential analysis of response rate. In addition, an intent-to -treat analysis was performed for all patients randomized to treatment. Another pr imary parameter in efficacy assessment was the change in the classification of the New York Heart Association as evaluated by the investigators. An end-point inferential analysis was performed by the Cochran-Mantel-Haenszel procedure. All comparisons were two-sided and were performed at the 5% level of significance . RESULTS OF PRIMARY EFFICACY CRITERIA

Figure I shows the mean change from baseline in exercise time after 6 weeks of treatment (a) and at the end of treatment (b) in the efficacysample and Fig . 2 at the end of treatment for all randomized patients. At week 6, there is a significant increase in exercise time from baseline in the quinapril 10- and 20-mg b.i.d . groups compared to the placebo group. The difference between the quinapril 5 mg b.i.d . group and the placebo group was not significant. At the end of treatment (Fig . I(bj), a further increase in exercise time compared to week 6 (Fig . ltaj) was only observed in the quinapril group receiving 20 mg b.i.d, At that time, there was a significant dose-related increase in exercise time between the quinapril groups compared to P 5 10 20 N=46 N=5 1 N=48 N=50 the placebo group. The intent-to-treat analysis (Fig. 2) confirmed these Figure 1 Mean changes in exercise time from baseline after 6 weeks results . Figure 3 shows the response rate at the end of of treatment (a) and at the end of treatment (b) in the efficacy sample. therapy. There were significant differences in response (Placebo vs. quinapril * P < 0'05; ** P < 0·0 I) . (increase in total exercise time of > 20% from baseline) for all quinapril groups compared to placebo, but not between groups on the ACE inhibitor. The same statistical were obtained for patients with baseline NYHA 111 than baseline NYHA 11 (Table 3). Changes in NYHA classifiresults were obtained in an intent-to-treat analysis. The analysis of the change from baseline in the classifi- cation are statistically significant between placebo- and cation according to the New York Heart Association for quinapril-treated patients combining the data for all all patients in the efficacy sample at the end of the study quinapril treatment groups (P=0 ·004). Changes in exercise time were analysed in some defined showed greater rates of improvement for all quinapril groups compared to placebo, but the overall statistical patient subgroups at the end of treatment. The subgroups test showed no significant treatment difference after an considered were patients with and without digitalis mediadjustment for baseline (P = 0·11). Greater improvements cation, patients with and without diuretic comedication,

G. A. J. Riegger

708

80

and patients only on quinapril monotherapy without any concomitant medications. Analysis of covariance for the effect of digitalis therapy was not significant. Stratification for diuretics as concomitant medication showed a greater effect in patients not treated with diuretics than in those who were on diuretics (P=0·0019) . The patients on quinapril as monotherapy were compared to the other treatment subgroups in a subgroup analysis. The increase in exercise time in the quinapril group alone is comparable to the diuretic group alone (from 268 to 307 s and from 285 to 330 s, respectively). The response rate in patients on quinapril as monotherapy was 46% compared to 25% in patients on diuretics alone.

**

70 60 V>

u

c;

50

0

o

'"

(fJ

40 30

20 10 0

RESULTS OF SECONDARY EFFICACY CRITERIA

-10 P N=5 6

5

N= 57

10

20

N= 56

N= 56

Figure 2 Mean changes in exercise time from baseline at the end of treatment for all randomized patients (intent-to-treat analysis). (Placebo vs. quinapril • P