2019 Volume 83 Issue 6 Pages 1269-1277
Background: The comparative tolerability, efficacy, and safety of bisoprolol and carvedilol have not been established in Japanese patients with heart failure and reduced ejection fraction (HFrEF).
Methods and Results: The CIBIS-J trial is a multicenter, open-label, non-inferiority randomized controlled trial of bisoprolol vs. carvedilol in 217 patients with HFrEF (EF ≤40%). The primary endpoint was tolerability, defined as reaching and maintaining the maximum maintenance dose (bisoprolol 5 mg/day or carvedilol 20 mg/day) during 48 weeks of treatment. The primary endpoint was achieved in 41.4% of patients in bisoprolol (n=111) and 42.5% in carvedilol (n=106) groups. The non-inferiority of tolerability of bisoprolol compared with carvedilol was not supported, however, neither β-blocker was superior with regard to tolerability. Heart rate (HR) decreased in both groups and its decrease from baseline was significantly greater in the bisoprolol group (20.3 vs. 15.4 beats/min at 24 week, P<0.05). Plasma B-type natriuretic peptide (BNP) levels decreased in both groups and the decrease was significantly greater in the carvedilol group (12.4 vs. 39.0 % at 24 weeks, P<0.05).
Conclusions: There were no significant differences between bisoprolol and carvedilol in the tolerability of target doses in Japanese HFrEF patients. The clinical efficacy and safety were also similar despite the greater reduction in HR by bisoprolol and plasma BNP by carvedilol.
Previous large-scale clinical trials have demonstrated that β-blockers, including carvedilol, bisoprolol and metoprolol succinate controlled release, can reduce mortality by 30% and hospitalizations by 40% in patients with heart failure and reduced ejection fraction (HFrEF).1–5 The use of maximum tolerable doses of these β-blockers is recommended for HFrEF patients according to the guidelines from ESC, ACC/AHA/HFSA, and JCS/JHFS.6–8 However, previous studies have shown that only 20–40% of these patients are taking β-blockers and their mean dose is half the recommended target dose.9,10
The underuse and underdosing of β-blockers may be caused by a lack of tolerability in patients, especially in those who are older and have multiple comorbidities, and drugs. β-blockers are not uniform in their pharmacological profiles, comprising selective β1-blockers, bisoprolol and metoprolol, and the nonselective β1- and β2-blocker, carvedilol. Therefore, tolerability may differ among these β-blockers, although COMET (Carvedilol Or Metoprolol European Trial) reported that overall tolerability did not differ between carvedilol and metoprolol.11 Similarly, CIBIS-ELD (Cardiac Insufficiency Bisoprolol Study in Elderly) reported that tolerability to target dose was comparable between bisoprolol and carvedilol.12 However, one needs to be cautious in applying these findings to Japanese patients, especially because the maximum recommended dose of β-blockers is much less in Japan compared with other countries: 5 vs. 10–20 mg/day for bisoprolol and 20 vs. 50–100 mg/day for carvedilol.6–8
In Japan, efficacy and safety were compared between bisoprolol and carvedilol in the randomized, controlled, double-blind trial, MAIN-CHF II (Multistep Administration of bisoprolol IN Chronic Heart Failure II) study, which showed that tolerability was similar between these drugs.13 However, the number of patients in this study was as small as 59 (31 for bisoprolol and 28 for carvedilol groups) and only 35 patients completed the study because it was prematurely terminated by the approval of bisoprolol for HF use by the government agency.
We thus designed a multicenter, open-label, randomized, parallel group comparative study, the CIBIS-J trial (Cardiac Insufficiency Bisoprolol Study in Japan), to investigate the tolerability, efficacy, and safety of β-blockers in Japanese HFrEF patients. The primary endpoint was defined as tolerability of bisoprolol and carvedilol when used at their guideline-recommended target doses, 5 and 20 mg/day, respectively.
CIBIS-J is an investigator-initiated, open-label, randomized, parallel group trial in Japanese patients with HFrEF. The hypothesis was that the tolerability of bisoprolol is not inferior to carvedilol when bisoprolol or carvedilol is given for 48 weeks to achieve the maximum maintenance dose as the primary endpoint. The trial was conducted at 71 study sites in Japan.
Inclusion and exclusion criteria are listed in Table 1. Patient recruitment started in June, 2013 and follow-up ended in July, 2016. The study protocol was approved by the Ethics Committee of Hokkaido University Hospital and of each institution, and registered with UMIN clinical Trial Registry; UMIN000011274. All patients provided written informed consent before enrollment.
| Inclusion criteria |
| • Age ≥20 to <85 years |
| • Chronic HF caused by IHD or dilated cardiomyopathy |
| • NYHA functional class ≥II |
| • LVEF ≤40% |
| • Treated with ACEI or ARB, diuretics, and digitalis |
| • Not receiving β-blocker (except for eye drops) within 8 weeks |
| • Written informed consent |
| Exclusion criteria |
| • Contraindication to bisoprolol or carvedilol |
| • AMI within 8 weeks |
| • Stroke or severe cerebrovascular disease within 1 year |
| • Malignancy with poor prognosis within 5 years |
| • Scheduled coronary artery bypass grafting or percutaneous coronary intervention during the study |
ACEI, angiotensin-converting enzyme inhibitor; AMI, acute myocardial infarction; ARB, angiotensin II receptor blocker; HF, heart failure; IHD, ischemic heart disease; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association.
Eligible patients were centrally randomized to the bisoprolol or carvedilol group in a 1:1 ratio. Stratified allocation was performed using the following factors: NYHA functional class (I/II or III/IV), heart rate (HR: <75 or ≥75 beats/min), age (<65 or ≥65 years), and inpatient or outpatient.
As shown in Figure 1, the study consisted of an observation phase (–8 weeks), titration phase (3–14 weeks), and the maintenance phase (34–45 weeks). The duration of each phase was set as the standard design and could be modified at each investigator’s discretion. During the titration phase, the dose was started at 0.625 mg of bisoprolol once daily or 1.25 mg of carvedilol twice daily and scheduled to be doubled at every visit to reach the target dose of 5 mg of bisoprolol or 20 mg of carvedilol within 3–14 weeks according to the titration scheme (Supplementary Figure 1). Investigators were free to delay titration or reduce the dose if clinically indicated. The titration phase was followed by the maintenance phase and the total period of follow-up was 48 weeks.
CIBIS-J study design consisted of observation, titration, and maintenance phases. Patients were randomized into the bisoprolol and carvedilol groups at 1:1. The dose was adjusted according to dose adjustment criteria at each step during the follow-up of 48 weeks. BID, twice daily; QD, once daily.
The primary endpoint was tolerability defined as achieving the target doses, 5 mg/day of bisoprolol or 20 mg/day of carvedilol, at Week 48 and maintained for at least 10 consecutive days. The secondary efficacy endpoints were changes in NYHA functional class, left ventricular EF (LVEF), LV end-diastolic volume (LVEDV)/LV end-systolic volume (LVESV), HR, and plasma B-type natriuretic peptide (BNP). The secondary safety endpoints included all-cause death; cardiovascular death; cardiovascular hospitalization; worsening HF defined as hospitalization because of worsening HF or intensification of treatment, including increasing the dose of diuretic, vasodilator, or inotropic agent or the addition of these drugs; HF death, including death from pump failure and worsening HF; sudden death including arrhythmia; and non-cardiovascular death. Safety endpoints also included the incidence of adverse events (AEs).
Statistical AnalysisTolerability and safety were evaluated in all patients taking at least 1 dose of the randomly assigned study drugs. The efficacy evaluation included all patients except those who did not have any data on secondary endpoints. The frequency distribution was compared using a chi-square test for nominal data, the Wilcoxon rank sum test for ordinal data, and a 2-sample t-test for quantitative values. Differences in the rate of achieving maximum maintenance dose (bisoprolol group−carvedilol group) and its 95% confidence interval (CI) between treatment groups (the primary endpoint) were calculated by Wald test. The non-inferiority of bisoprolol to carvedilol would be concluded when the lower limit of 95% CI was ≥−10% for the difference between groups. Based on previous studies, including MAIN-CHF II10 and CIBIS-ELD,7 the rate of achieving maximum maintenance dose for bisoprolol and carvedilol was assumed to be 35% and 25%, respectively. To verify the non-inferiority of bisoprolol to carvedilol on differences between groups, 83 patients per group were required with one-sided α level=2.5% and a power of 80%. Target sample size was thus determined as 110 patients per group. For the secondary endpoints, the changes at Weeks 24 and 48 from baseline were compared between groups. Time to the event in each group was obtained using Kaplan-Meier analyses. Comparison between groups was conducted by stratified log-rank test with adjustment factors for allocation. AEs were coded by preferred term (PT) and grouped by system organ class (SOC) using MedDRA/J version 17.1.
The level of significance of the test was two-sided 0.05 unless otherwise specified, using two-sided 95% CI. Statistical analysis was performed using SAS software (SAS for Windows Release ver. 9.2 or later, SAS Institute Inc.).
A total of 232 patients were randomly assigned to the bisoprolol group (116 patients) or the carvedilol group (116 patients). Of them, 217 patients (111 patients in the bisoprolol group and 106 patients in the carvedilol group) were included in the tolerability and safety analysis. The efficacy analysis was performed in 201 patients (105 patients in the bisoprolol group and 96 patients in the carvedilol group) (Figure 2).
Patient flowchart.
The patients’ baseline characteristics are shown in Table 2. Overall, there were no imbalances between treatment groups except for a higher prevalence of anemia in the bisoprolol group (25.2% vs. 14.2%, P=0.0408). The prevalence of anemia defined by hemoglobin levels also tended to be higher in the bisoprolol group, but did not reach statistical significance. The use of diabetic drugs, including sulfonylureas and DPP-4 inhibitors, were higher in the bisoprolol group.
| Bisoprolol (n=111) |
Carvedilol (n=106) |
P value | |
|---|---|---|---|
| Age (years) | |||
| Mean (SD) | 60.9 (14.3) | 59.9 (14.2) | 0.5879 |
| ≥65 | 47 (42.3) | 44 (41.5) | 0.9011 |
| Sex | |||
| Male | 85 (76.6) | 84 (79.2) | 0.6359 |
| BMI (kg/m2) | |||
| Mean (SD) | 23.8 (4.3) | 24.9 (5.6) | 0.0963 |
| ≥25 | 41 (37.6) | 43 (40.6) | 0.6574 |
| NYHA functional class | 0.9990 | ||
| I | 1 (0.9) | 1 (0.9) | |
| II | 73 (65.8) | 69 (65.1) | |
| III | 31 (27.9) | 32 (30.2) | |
| IV | 6 (5.4) | 4 (3.8) | |
| Etiology of HF | 0.7687 | ||
| IHD | 18 (16.2) | 18 (17.0) | |
| DCM | 71 (64.0) | 71 (67.0) | |
| Other | 22 (19.8) | 17 (16.0) | |
| HR (beats/min) | |||
| Mean (SD) | 85.0 (18.3) | 84.7 (17.7) | 0.8780 |
| ≥75 | 76 (68.5) | 69 (65.1) | 0.5977 |
| SBP (mmHg) | |||
| Mean (SD) | 120.7 (18.0) | 121.0 (17.8) | 0.9020 |
| DBP (mmHg) | |||
| Mean (SD) | 73.7 (13.3) | 75.2 (15.1) | 0.4441 |
| LVEF (%) | |||
| Mean (SD) | 28.8 (7.0) | 28.8 (7.8) | 0.9487 |
| Plasma BNP (pg/dL) | 0.9874 | ||
| Geometric mean (SD) | 169.92 (2.98) | 170.31 (2.95) | |
| Median (min, max) | 184.0 (6.8, 2,260.0) |
176.0 (12.2, 1,960.0) |
|
| Hemoglobin | |||
| Mean (SD) | 13.94 (2.291) | 14.51 (2.161) | 0.0593 |
| Anemia (defined by Hb levela) | 34 (30.9) | 21 (19.8) | 0.0613 |
| eGFR (mL/min/1.73 m2) | |||
| Mean (SD) | 61.50 (19.54) | 61.20 (18.93) | 0.9101 |
| <60 | 50 (45.0) | 51 (48.1) | 0.6506 |
| FEV1 (L) | |||
| Mean (SD) | 2.32 (0.84) | 2.38 (0.71) | 0.5499 |
| %VC (%) | |||
| Mean (SD) | 87.40 (18.94) | 89.29 (18.81) | 0.4654 |
| Medical history | |||
| Hypertension | 82 (73.9) | 68 (64.2) | 0.1212 |
| Diabetes mellitus | 42 (37.8) | 36 (34.0) | 0.5520 |
| Dyslipidemia | 44 (39.6) | 53 (50.0) | 0.1249 |
| Hyperuricemia | 60 (54.1) | 58 (54.7) | 0.9219 |
| Angina | 13 (11.7) | 12 (11.3) | 0.9282 |
| Myocardial infarction | 11 (9.9) | 11 (10.4) | 0.9092 |
| Atrial fibrillation | 27 (24.3) | 25 (23.6) | 0.8985 |
| Stroke | 4 (3.6) | 6 (5.7) | 0.4701 |
| COPD | 3 (2.7) | 5 (4.7) | 0.4312 |
| Bronchial asthma | 2 (1.8) | 1 (0.9) | 0.5883 |
| Renal failure | 13 (11.7) | 13 (12.3) | 0.9003 |
| Anemia | 28 (25.2) | 15 (14.2) | 0.0408 |
| Medications | |||
| ACEI | 52 (46.8) | 59 (55.7) | 0.1942 |
| ARB | 44 (39.6) | 35 (33.0) | 0.3110 |
| ACEI or ARB | 95 (85.6) | 94 (88.7) | 0.4968 |
| Mineralocorticoid-receptor antagonist | 46 (41.4) | 42 (39.6) | 0.7850 |
| Diuretic | 90 (81.1) | 87 (82.1) | 0.8502 |
| Digitalis | 14 (12.6) | 9 (8.5) | 0.3241 |
| Positive inotrope | 2 (1.8) | 2 (1.9) | 0.9629 |
| CCB | 18 (16.2) | 17 (16.0) | 0.9715 |
| Vasodilator | 6 (5.4) | 5 (4.7) | 0.8172 |
| Amiodarone | 6 (5.4) | 5 (4.7) | 0.8172 |
| Statin | 26 (23.4) | 28 (26.4) | 0.6104 |
| Lipid-lowering drug (other than statin) | 6 (5.4) | 3 (2.8) | 0.3416 |
| Aspirin | 15 (13.5) | 14 (13.2) | 0.9472 |
| Antiplatelet (other than aspirin) | 15 (13.5) | 9 (8.5) | 0.2383 |
| Anticoagulant | 33 (29.7) | 30 (28.3) | 0.8168 |
| Sulfonylurea | 9 (8.1) | 2 (1.9) | 0.0368 |
| Thiazolidinedione | 0 (0.0) | 3 (2.8) | 0.0743 |
| Biguanide | 5 (4.5) | 7 (6.6) | 0.4988 |
| DPP-4 inhibitor | 24 (21.6) | 12 (11.3) | 0.0414 |
| Insulin | 6 (5.4) | 6 (5.7) | 0.9345 |
Data are number of patients (%) or mean (SD; standard deviation). aDefined as hemoglobin level <12 g/dL in females and <13 g/mL in males. BMI, body mass index; BNP, B-type natriuretic peptide; CCB, calcium-channel blocker; DBP, diastolic blood pressure; DCM, dilated cardiomyopathy; DPP-4, dipeptidyl peptidase-4; eGFR, estimated glomerular filtration rate; FEV1, forced expiratory volume in 1 s; GLP-1, glucagon-like peptide-1; HR, heart rate; SBP, systolic blood pressure; %VC, % vital capacity. Other abbreviations as in Table 1.
The percentage of achieving maximum maintenance dose (primary endpoint) was 41.4% in the bisoprolol group (95% CI: 32.2%, 51.2%) and 42.5% in the carvedilol group (95% CI: 32.9%, 52.4%) (Figure 3). The non-inferiority was not significantly (P=0.0899) supported based on the 95% CI of the difference (−14.1%, 12.1%), less than its prespecified lower limit (−10%). However, by chi-square test there was no significant difference between the bisoprolol and carvedilol groups in the percentage value of achieving maximum maintenance dose (P=0.8800).
Primary endpoint (tolerability), defined as achieving the target doses, 5 mg/day of bisoprolol or 20 mg/day of carvedilol, at Week 48 and maintained for at least 10 consecutive days. CI, confidence interval.
The doses achieved at Weeks 0, 24, and 48 in the bisoprolol and carvedilol groups are shown in Figure 4. The mean daily dose at Week 48 was 3.53 mg for bisoprolol and 14.13 mg for carvedilol. The proportion of patients receiving the drug at greater than half the target dose was 69.5% and 61.0% in the bisoprolol and carvedilol groups at Week 24 and 81.4% and 75.0% at Week 48, respectively. It tended to be higher in the bisoprolol group than in the carvedilol group, but did not reach statistical significance.
Percentage of dose achieved at baseline, Week 24, and Week 48 in the bisoprolol and carvedilol groups.
We performed post-hoc subgroup analysis specified by age, sex, BMI, etiology, and atrial fibrillation. Tolerability, the primary endpoint, was lower in the subgroups of elderly, lower BMI, ischemic etiology, and the presence of atrial fibrillation for both β-blockers (Supplementary Table 1). Moreover, the primary endpoint did not differ between subgroups.
The proportion of patients achieving 50% of maximum maintenance dose increased to 74.5–79.3% and for 25% was further increased to 93.4–95.5% (Supplementary Figure 2). Furthermore, the tolerability of bisoprolol 1.25 mg and carvedilol 5 mg was as high as 100% (n=7) and 80.0% (n=5), respectively, for elderly patients with ischemic cardiomyopathy whereas that for bisoprolol 5 mg and carvedilol 20 mg was 45.9% (n=61) and 48.3% (n=60), respectively, for younger patients with dilated cardiomyopathy (Supplementary Figure 2). Tolerability rose when the maintenance dose was set to 25% of the maximum dose specifically for elderly patients with ischemic cardiomyopathy.
Secondary Endpoints (Table 3)NYHA functional class improved significantly to the same extent at Weeks 24 and 48 compared with baseline in each group (P<0.001) (Supplementary Figure 3). HR decreased significantly at Weeks 24 and 48 compared with baseline in each treatment group, but the extent was greater in the bisoprolol group than in the carvedilol group (−20.3 vs. −15.4 beats/min at Week 24, P=0.0315; −19.7 vs. −16.5 beats/min at Week 48, P=0.0796) (Supplementary Figure 4). Systolic blood pressure (SBP) was slightly increased by approximately 4 mmHg and diastolic BP (DBP) decreased by 1 mmHg in both groups. These changes in SBP and DBP did not differ between groups. Mean LVEF increased from baseline in both groups (Supplementary Figure 5). The increase in LVEF was 15.0% and 17.7% in the bisoprolol and carvedilol groups, respectively, at Week 48. The estimated difference did not differ between groups. Mean LVEDV and LVESV decreased from baseline in both groups and these changes were also similar (Supplementary Figure 5). Plasma BNP levels decreased from baseline in both groups, with a greater decrease in the carvedilol group at Week 24 but no significant differences at Week 48 (Supplementary Figure 6).
| Bisoprolol | Carvedilol | Intergroup difference (bisoprolol-carvedilol) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Week 24 | Week 48 | |||||||||
| Baseline | Week 24 | Week 48 | Baseline | Week 24 | Week 48 | Difference (95% Cl) |
P value | Difference (95% Cl) |
P value | |
| NYHA functional class n, (%) |
0.5636 | 0.4729 | ||||||||
| I | 0 (0.0) | 41 (42.7) | 44 (51.2) | 0 (0.0) | 44 (49.4) | 49 (58.3) | ||||
| II | 66 (68.8) | 52 (54.2) | 39 (45.3) | 60 (67.4) | 43 (48.3) | 32 (38.1) | ||||
| III | 24 (25.0) | 3 (3.1) | 3 (3.5) | 25 (28.1) | 2 (2.2) | 3 (3.6) | ||||
| IV | 6 (6.3) | 0 (0.0) | 0 (0.0) | 4 (4.5) | 0 (0.0) | 0 (0.0) | ||||
| HR (beats/min) | ||||||||||
| Mean (SD) | 86.4 (17.8) |
66.1 (12.1) |
66.6 (14.3) |
84.4 (18.3) |
69.5 (13.4) |
70.1 (12.6) |
||||
| Change (95% CI) | −20.3 (−23.7, −16.8) |
−19.7 (−23.6, −15.8) |
−15.4 (−19.4, −11.4) |
−16.5 (−20.3, −12.7) |
−3.8 (−7.3, −0.3) |
0.0315 | −3.4 (−7.2, 0.4) |
0.0796 | ||
| HR <50 n, (%)a | 0/105 (0.0) |
1/95 (1.1) |
5/76 (6.6) |
1/96 (1.0) |
1/82 (1.2) |
2/78 (2.6) |
||||
| SBP (mmHg) | ||||||||||
| Mean (SD) | 120.9 (17.9) |
124.6 (19.4) |
124.0 (19.5) |
121.1 (18.1) |
127.1 (18.3) |
125.0 (20.6) |
||||
| Change (95% CI) | 4.2 (0.0, 8.4) |
3.6 (−0.8, 8.1) |
5.9 (0.9, 10.8) |
4.3 (−0.8, 9.3) |
−2.3 (−7.7, 3.0) |
0.3925 | −0.9 (−6.6, 4.9) |
0.7706 | ||
| DBP (mmHg) | ||||||||||
| Mean (SD) | 73.6 (13.3) |
72.8 (13.3) |
73.0 (11.1) |
75.2 (15.5) |
75.4 (13.5) |
73.8 (14.1) |
||||
| Change (95% CI) | −0.8 (−3.6, 1.9) |
−1.3 (−4.1, 1.5) |
−0.0 (−3.4, 3.3) |
−0.3 (−4.0, 3.4) |
−2.0 (−5.5, 1.5) |
0.260 3 |
−0.9 (−4.5, 2.7) |
0.6259 | ||
| LVEF (%) | ||||||||||
| Mean (SD) | 28.4 (6.9) |
42.6 (12.3) |
43.4 (12.6) |
28.9 (7.8) |
44.5 (12.3) |
46.3 (11.4) |
||||
| Change (95% CI) | 14.6 (11.9, 17.3) |
15.0 (12.0, 18.0) |
15.7 (13.0, 18.4) |
17.7 (15.0, 20.5) |
−1.5 (−5.1, 2.19) |
0.4172 | −2.8 (−6.6, 1.0) |
0.1412 | ||
| LVEDV (mL) | ||||||||||
| Mean (SD) | 185.4 (81.1) |
157.4 (111.1) |
143.2 (69.9) |
172.1 (55.8) |
135.6 (53.4) |
129.8 (55.1) |
||||
| Change (95% CI) | −35.2 (−58.4, −12.0) |
−45.0 (−60.5, −29.6) |
−35.8 (−46.2, −25.4) |
−44.1 (−55.1, −33.1) |
10.4 (−14.3, 35.1) |
0.4063 | 5.3 (−10.5, 21.1) |
0.5074 | ||
| LVESV (mL) | ||||||||||
| Mean (SD) | 131.9 (66.4) |
99.6 (109.2) |
86.1 (58.1) |
123.4 (48.5) |
78.7 (45.0) |
73.0 (47.3) |
||||
| Change (95% CI) | −38.2 (−62.3, −14.0) |
−47.4 (−60.7, −34.6) |
−44.1 (−52.3, −35.8) |
−52.0 (−61.0, −42.9) |
14.3 (−10.7, 39.2) |
0.2607 | 8.1 (−5.2, 21.3) |
0.2290 | ||
| BNP (pg/mL) | ||||||||||
| Geometric mean (geometric SD) |
170.40 (3.03) |
85.71 (3.99) |
63.80 (4.68) |
166.02 (2.87) |
48.05 (4.50) |
48.72 (4.98) |
||||
| % change (95% CI) | −12.42 (79.35) |
−12.49 (95.19) |
−38.98 (73.04) |
−37.32 (68.12) |
26.56 (3.26, 49.87) |
0.0257 | 25.23 (−2.17, 52.63) |
0.0709 | ||
aProportion of patients achieving HR <50 beats/min. CI, confidence interval; %FS, fractional shortening; LVEDV, left ventricular end-diastolic volume; LVESV, left ventricular endsystolic volume; SD, standard deviation. Other abbreviations as in Tables 1,2.
The incidence of cardiovascular events, including death and hospitalization, were similar between groups during the follow-up (Table 4). There were no cardiovascular deaths. Hospitalization for cardiovascular causes was also similar between groups. The most prevalent events were worsening HF and this was also similar between groups. The incidence of other AEs was also similar between groups. HF reported as an AE by the investigators tended to be higher in the bisoprolol group than in the carvedilol group (16.2% vs. 7.5%, P=0.0602, Supplementary Table 2), but was not observed as worsening HF as a secondary efficacy endpoint (12.6 vs. 7.5%, P=0.2636, Table 4).
| Events | Bisoprolol [n=111] | Carvediol [n=106] | P value |
|---|---|---|---|
| n (%) | n (%) | ||
| All-cause death | 4 (3.6) | 3 (2.8) | 1.0000 |
| CV death | 0 (0.0) | 0 (0.0) | – |
| CV hospitalization | 8 (7.2) | 9 (8.5) | 0.8034 |
| Worsening HFa | 14 (12.6) | 8 (7.5) | 0.2636 |
| HF deathb | 0 (0.0) | 0 (0.0) | – |
| Sudden death | 0 (0.0) | 0 (0.0) | – |
| Non-CV death | 4 (3.6) | 3 (2.8) | 1.0000 |
aHospitalization or intensification of HF treatment for worsening of HF. bDeath from pump failure or worsening HF. CV, cardiovascular; HF, heart failure.
A summary of AEs is shown in Supplementary Table 2. The incidence of AEs was 37.8% and 32.1% in the bisoprolol group and carvedilol group, respectively. Commonly observed AEs by SOC were cardiac disorders (23.4% vs. 10.4%) and HF by PT (16.2% vs. 7.5%). AEs (side effects) at least possibly related to the study treatment were observed in 10.8% and 8.5%, serious AEs in 22.5% and 20.8%, and AEs leading to dose reduction or discontinuation of treatment in 9.9% and 7.5% in bisoprolol and carvedilol groups, respectively. Clinically significant abnormal laboratory values were not observed.
The present study demonstrated that the proportions of patients reaching the target doses did not differ between bisoprolol and carvedilol in Japanese HFrEF patients. The improvements in NYHA functional class and LVEF were also similar, despite the greater reduction in HR by bisoprolol and of plasma BNP by carvedilol.
TolerabilityThe proportions of patients reaching the maximum dose were similar between the bisoprolol and carvedilol groups (41.4% vs. 42.5 %), even though non-inferiority was not confirmed based on the statistical analysis (Figure 3). The majority of patients (75–80%) received the study drug at greater than half the target dose (Figure 4). These findings were consistent with those from CIBIS-ELD, demonstrating the comparable tolerability of bisoprolol and carvedilol, in which dosing was achieved by forced titration.12 Similar results regarding the tolerability of β-blockers were also reported by observational studies from Europe.14,15 A European survey showed that the target doses of carvedilol, bisoprolol, and metoprolol, as defined by the ESC guidelines, were reached in 37%, 21%, and 21% of patients, respectively.14 In another study of primary care from Scotland, β-blockers were reported to be used in 62% and target doses were reached in 34%.15 Based on those studies, the proportion of patients who can be titrated to target doses of β-blockers in clinical practice is considered to be in the range of 20–40%. In contrast, these values are lower than those reported from randomized controlled trials, in which 40–80% of patients reached the recommended target doses.16,17 However, even in those trials, a considerable number of patients could not reach the target dose at the end of the titration phase.18 In addition, the doses achieved during the titration phase could not be maintained and had to be reduced or interrupted because of intolerance during the maintenance phase.
Despite the low tolerability to β-blockers of around 40–50% shown in this study, it is important to uptitrate β-blockers to the maximum tolerated dose according to the recommendations of guidelines.6–8 However, it is also reasonable to adjust the doses according to the patient’s age and etiology of HF.
EfficacyNYHA functional class improved in both treatment groups and there were no between-group differences (Table 3, Supplementary Figure 3) in agreement with previous studies, including from Japan.13
HR was substantially reduced in both groups and its reduction was greater in the bisoprolol group than in the carvedilol group (Table 3, Supplementary Figure 4). This might be caused by the relatively higher proportion of patients receiving the drug at greater than half the target dose in the bisoprolol group than in the carvedilol group (69.5% vs. 61.0% at Week 24; 81.4% vs. 75.0% at Week 48) (Figure 4). In addition, pharmacological differences might also explain these results, based on the fact that sole α-blockade can increase HR and the combination of α- and β-blockade by carvedilol might weaken the HR-lowering effect of β-blockers. HR reduction is one of the most important therapeutic effects of β-blockers on mortality risk in patients with HF. A meta-analysis of β-blocker trials reported that every reduction of 5 beats/min in HR reduced the mortality rate by 18%.19 Importantly, the reduction in HR by bisoprolol or carvedilol at the doses used in this study was similar to that obtained by the higher dose used in randomized controlled trials performed in Western countries, suggesting that similar mortality benefit may be expected in Japanese HFrEF patients. SBP and DBP did not decrease, but SBP rather increased (Table 3, Supplementary Figure 4), consistent with previous trials, including the MAIN-CHF II study.13,20,21
LVEF increased in both groups (Table 3, Supplementary Figure 5) and this increase was similar to that reported in other previous studies in Japan, including the MAIN-CHF II13 and MUCHA22 studies. It was also consistent with previous reports from other countries.23 Thus, administration of bisoprolol or carvedilol at the doses used in this study increased LVEF as much as did the larger dose used in Western countries. LVEDV and LVESV substantially decreased from baseline in both groups (Table 3, Supplementary Figure 5). The changes were similar between bisoprolol and carvedilol. These results are also consistent with those reported by previous studies that examined the correlation between ventricular remodeling and mortality.23 Ventricular remodeling, defined as ventricular enlargement and dysfunction, is regarded as a predictor of mortality risk in patients with HF.24 Therefore, the effects of β-blockers on ventricular remodeling shown in the present study might be associated with their efficacy in clinical outcomes.
Plasma BNP levels decreased below baseline in both groups (Table 3, Supplementary Figure 6) and the changes may be attributed to subsequent improvements in cardiac function under sustained β-blocker treatment. Interestingly, this decrease was significantly greater in the carvedilol group at Week 24, but was not significant at Week 48 (Table 3, Supplementary Figure 6). These changes might be related to differences in mechanical cardiac loading between bisoprolol and carvedilol during the titration phase, which has been shown in a previous trial.13 Subsequent improvements in cardiac function might diminish this difference between groups at Week 48.
Study LimitationsThere are several potential limitations of this study to be acknowledged. First, the CIBIS -J study was an open-label trial, which might lead to a potential bias in titration and maintenance of the study drug. However, the percentage values for achieving the primary endpoint were even higher in CIBIS-J than in CIBIS-ELD, a randomized parallel dummy, double-blind trial (40% vs. 25%). In addition, investigators were free to delay titration or reduce the dose if clinically indicated in either group. Therefore, it seems unlikely that the trial design would affect the study results. Second, CIBIS-J did not document reasons for failing to achieve the target dose during the titration phase. Potential reasons for downtitration, slowed titration, or discontinuation of β-blockers might include undesirable reduction in BP or HR. When β-blockers cannot be uptitrated as usual, their maintenance doses, even at a half or quarter of the maximum dose, might be suitable for patients. Patients in the bisoprolol group were more likely to have a greater reduction in HR as shown in Table 3. Therefore, these reasons might differ between groups. Third, a correlation between tolerability to the target dose or titration success and clinical outcomes including death or hospitalization could not be investigated because of the small number of cardiovascular events in this study (Table 4) and the short follow-up (48 weeks). However, decreases in HR and LV volumes comparable to those reported in earlier large-scale randomized clinical trials might suggest similar improvements in outcomes in our patients.19,23
There were no significant differences between bisoprolol and carvedilol in tolerability of the target doses by Japanese HFrEF patients. The clinical efficacy and safety were also similar, despite a greater reduction in HR by bisoprolol and of plasma BNP by carvedilol.
The authors thank the study participants for their time and commitment and the investigators who made this study possible (see Supplementary Appendix).
This study was funded by Mitsubishi Tanabe Pharma Corporation (Osaka, Japan). The study funder gave information for authors in the preparation for study drafting, but did not participate in the study plan, data collection, data analysis, interpretation, or writing the manuscript. The authors made the final decision on the study plan and performed it as an investigator-initiated clinical study. The data were collected and analyzed by the contract research organization Mebix (Tokyo, Japan) according to the protocol. The manuscript was fully reviewed and revised by the authors. The authors made the final decision on submission of the article.
H.T. received consultancy fees from Novartis Pharma K.K, Pfizer Japan Inc., Bayer Yakuhin, Ltd., Nippon Boehringer Ingelheim Co., Ltd., and Ono Pharmaceutical Co., Ltd.; received the speakers’ bureau and/or honoraria from Daiichi Sankyo Co., Ltd., MSD K.K., Mitsubishi Tanabe Pharma Corp., Teijin Pharma Ltd., Bristol-Myers Squibb Company, Takeda Pharmaceutical Co., Ltd., Nippon Boehringer, Ingelheim Co., Ltd., and Bayer Yakuhin, Ltd.; received research funds from Takeda Pharmaceutical Co., Ltd., Bayer Yakuhin, Ltd., Nippon Boehringer Ingelheim Co., Ltd., Mitsubishi Tanabe Pharma Corp., Sanofi K.K., and Daiichi Sankyo Co., Ltd.; and is affiliated with an endowed department sponsored by Acterion Pharmaceuticals Japan Ltd. S.M. received remuneration for lectures from Nippon Boehringer Ingelheim Co., Ltd., Bayer Yakuhin, Ltd., and Otsuka Pharmaceutical Co., Ltd.; and received scholarship funds from Medtronic Japan Co., Ltd. T. Masuyama received remuneration for lectures from Otsuka Pharmaceutical Co., Ltd.; received scholarship funds from Actelion Pharmaceuticals Japan Ltd., Astellas Pharma Inc., MSD K.K., Otsuka Pharmaceutical Co., Ltd., Ono Pharmaceutical Co., Ltd., Kowa Pharmaceutical Co., Ltd., Daiichi Sankyo Co., Ltd., Takeda Pharmaceutical Co., Ltd., Mitsubishi Tanabe Pharma Corp., Teijin Pharma Ltd., Nippon Shinyaku Co., Ltd., Nippon Boehringer Ingelheim Co., Ltd., Bayer Yakuhin, Ltd., and Pfizer Japan Inc.; and is affiliated with a department endowed by Abbott Medical Japan Co., Ltd., Medtronic Japan Co., Ltd., and Nippon Boehringer Ingelheim Co., Ltd. Y.S. received remuneration for lectures from Novartis Pharma K.K., Mitsubishi Tanabe Pharma Corp., Otsuka Pharmaceutical Co., Ltd., Daiichi Sankyo Co., Ltd., Pfizer Japan Inc., and Nippon Boehringer Ingelheim Co., Ltd.; received trust research/joint research funds from Novartis Pharma K.K., Ono Pharmaceutical Co., Ltd., St. Jude Medical Japan Co., Ltd., Bayer Holding Ltd., and Terumo Corp.; received scholarship funds from Daiichi Sankyo Co., Ltd., Otsuka Pharmaceutical Co., Ltd., Kyowa Hakko Kirin Co., Ltd., Sumitomo Dainippon Pharma Co., Ltd., Astellas Pharma Inc., Takeda Pharmaceutical Co., Ltd., Teijin Pharma Ltd., Mitsubishi Tanabe Pharma Corp., Shionogi & Co., Ltd, Kowa Pharmaceutical Co., Ltd., and Actelion Pharmaceuticals Japan Ltd.; and is affiliated with a department endowed by MSD K.K. I.K. received remuneration for lectures from MSD K.K., Daiichi Sankyo Co., Ltd., Takeda Pharmaceutical Co., Ltd., Ono Pharmaceutical Co., Ltd., Nippon Boehringer Ingelheim Co., Ltd., Pfizer Japan Inc., and Toa Eiyo Ltd.; received research funds from Ono Pharmaceutical Co. Ltd.; and received scholarship funds from Astellas Pharma Inc., Edwards Lifesciences Corp., Otsuka Pharmaceutical Co. Ltd., Kowa Pharmaceutical Co. Ltd., Daiichi Sankyo Co., Ltd., Sumitomo Dainippon Pharma Co., Ltd., Takeda Pharmaceutical Co., Ltd., Mitsubishi Tanabe Pharma Corp., Teijin Pharma Ltd., Toa Eiyo Ltd., Nipro Corp., and Terumo Corp. T. Murohara received remuneration for lectures from Daiichi Sankyo Co., Ltd. and Mitsubishi Tanabe Pharma Corp.; and received scholarship funds from Daiichi Sankyo Co., Ltd. and Mitsubishi Tanabe Pharma Corp. S.K. has no conflicts of interest.
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-18-1199
