FormalPara Key Points

We undertook an interim analysis of an observational study in 655 Chinese patients who were prescribed ivabradine for the treatment of chronic heart failure.

Over 6 months, ivabradine was associated with significant reductions in heart rate and significant improvements in New York Heart Association functional class and quality of life.

Ivabradine was very well tolerated, with <10% of patients experiencing treatment-related adverse events.

1 Introduction

In patients with chronic heart failure (HF), compensatory increases in heart rate (HR) not only maintain adequate cardiac output but also increase the risk of disease progression. Persistent elevations in HR can lead to decreased myocardial contractility [1], increased ventricular afterload [2], reduced myocardial perfusion [3], and increased myocardial oxygen consumption [4]. These factors lead to a progressive decline in cardiac function, recurrent dyspnea and peripheral edema, increased levels of B-type natriuretic peptide (BNP), and myocardial remodeling [5]. In turn, worsening HF leads to further increases in HR via activation of the neurohumoral system, creating a self-propagating cycle of events that increases the risk of hospitalization and death.

Data from observational studies [6, 7] and SHIFT (Systolic Heart failure treatment with the If inhibitor ivabradine Trial) [8] have demonstrated that increased HR is a predictor of short- and long-term mortality in patients with HF. Consequently, HR has become an established therapeutic target in guidelines on HF treatment [9,10,11,12]. In patients with HF with reduced ejection fraction (HFrEF), reductions in HR lead to improvements in cardiac function; this in turn leads to a reduction in symptoms and improved outcomes [13, 14].

Ivabradine is a novel treatment for HF that reduces HR by selectively inhibiting the If current in the sinoatrial node [10]. In 2010, the SHIFT study demonstrated that ivabradine, added to standard therapy, significantly reduced the risk of hospitalization due to worsening HF in patients with HFrEF, compared with placebo [14]. Subsequently, ivabradine treatment was incorporated into clinical guidelines for treatment intensification in patients who, despite guideline-directed HF treatment including an optimized β-blocker, have symptomatic HFrEF (New York Heart Association [NYHA] class II or III, with left ventricular ejection fraction [LVEF] ≤ 35%) and a resting HR of ≥ 70 beats per minute (bpm) in sinus rhythm [11].

Observational studies of ivabradine conducted in Germany (INTENSIFY [15] and RELIf-CHF [16]) and the UK (LIVE:LIFE [17]) have confirmed that ivabradine improves symptoms and health-related quality of life (QoL) in patients with HF in real-world clinical practice. However, to date, there is little published real-world evidence to support the use of ivabradine in Chinese patients with HF. We therefore designed an observational study to assess the effectiveness and safety of ivabradine in this population.

The POSITIVE (POSt-authorization drug Intensive surveillance monitoring sTudy of IVabradinE in patients with HF) study is the first major observational study of ivabradine to be undertaken in China, and indeed is the first large study of ivabradine to assess real-world outcomes of HF in China. Its purpose is to assess the effectiveness and safety of ivabradine in Chinese patients with HFrEF. In this article, we present the findings of an interim analysis of the POSITIVE database.

2 Methods

The POSITIVE study is an ongoing, prospective, observational study of patients with HFrEF receiving care at one of 63 hospitals in China (see Appendix for full list). To be included, patients needed to have an LVEF ≤ 40% in the previous 3 months; NYHA class II, III or IV HF; and a resting HR ≥ 75 bpm in sinus rhythm. In addition, patients had to be receiving standard treatment for HF (β-blockers, or other agents if β-blockers were not tolerated or were contraindicated); meet the indication criteria for ivabradine in China; and to have been prescribed ivabradine by their treating physician.

According to prescribing guidelines in China, the starting dose of ivabradine is 5 mg twice daily, or 2.5 mg twice daily in patients aged > 75 years. The dosage can subsequently be increased if necessary, based on HR measurement, to a maximum of 7.5 mg twice daily. In the current study, investigators could adjust the dosage based on the patient’s condition and according to the dosage recommendations in the prescribing information. Patients with a resting HR of 50–60 bpm continued to receive the starting dose, whereas patients with a resting HR that was persistently > 60 bpm could have the ivabradine dose increased up to a maximum of 7.5 mg twice daily. The ivabradine dose was decreased or treatment discontinued if the patient’s resting HR was persistently < 50 bpm or if they had symptoms of bradycardia.

As per usual clinical practice for the monitoring of patients receiving care for HFrEF, patients had monthly follow-up visits, during which the ivabradine dosage could be adjusted. In this observational study, data on clinical status were collected at two follow-up visits, at 1 month (M1) and 6 months (M6) after inclusion. If patients were unable to attend the hospital for follow-up, they were interviewed by telephone. At baseline and at each follow-up visit, patients underwent a physical examination and blood was taken for routine laboratory analysis. NYHA class was assessed and the validated Chinese-language version of Kansas City Cardiomyopathy Questionnaire (KCCQ) [18] was administered to monitor changes in health-related QoL over time. In the KCCQ, 23 items are rated and scaled to generate a score of between 0 and 100, where lower scores indicate poor QoL and high scores indicate good QoL. The minimum change in KCCQ score that is considered clinically meaningful is 5 points [19]. Patients were also requested to keep a diary of any adverse events (AEs) and were asked about these AEs during telephone follow-up. Bradycardia that was symptomatic or characterized by an HR of < 40 bpm was recorded as an AE.

The primary outcome of interest was the change in HR between baseline and each of the follow-up visits. Secondary outcomes were the changes in NYHA functional class and KCCQ scores, as well as safety and tolerability. Changes in blood BNP/N-terminal pro-BNP (NT-proBNP) levels and LVEF were also assessed.

Informed consent to participate in this study was obtained from all participants. The study was approved by the Ethics Committee of Zhongshan Hospital Fudan University, and also underwent ethical review and approval at all participating centers. The trial was registered on the ISRCTN database (registration number: 11703380) on November 8, 2016.

An interim analysis was planned for when ≥500 patients had been enrolled and had completed two follow-up visits. In practice, we applied a cut-off date of October 31, 2019, and analyzed data for all patients enrolled up to that point.

Statistical analysis was performed using SAS version 9.1.3 (SAS Institute, Inc., Cary, NC, USA). The full analysis set (FAS) included all enrolled patients who provided informed consent and received at least one dose of ivabradine. The efficacy population consisted of the FAS population who took ivabradine for at least 1 month (i.e., excluding those who discontinued or changed ivabradine at the M1 visit or prematurely discontinued prior to the M1 visit). The primary efficacy analysis was carried out in the efficacy population. Unless otherwise specified, analysis of patient demographics, secondary efficacy endpoints and safety endpoints were carried out in the FAS. For continuous variables, either the mean and standard deviation (SD) or the median and interquartile range (IQR) were calculated. Categorical variables were summarized using frequency, which was expressed as an absolute number (n) and percentage (%). To analyze differences in outcomes of interest between follow-up visits and baseline, we used either the paired Student’s t-test or the Wilcoxon signed-rank test, depending on the distribution of the data; for ranked variables (such as NYHA class), we used the Wilcoxon signed-rank test. A p-value <0.05 was considered to be statistically significant.

3 Results

3.1 Baseline Characteristics

At October 31, 2019, 655 patients had been enrolled in the study and were included in the FAS interim analysis. Of these, 554 (84.6%) and 449 (68.5%) subjects had completed the M1 and M6 follow-up visits, respectively. The efficacy analysis included 534 patients; within this group, baseline ECG HR data were missing for 23 patients, and baseline KCCQ data were missing for 37 patients. Baseline NYHA class data were available for all 534 patients in the efficacy population. Data were missing at M1 and M6 for 306 and 393 patients, respectively, for HR; 122 and 227 patients, respectively, for NYHA class; and 156 and 253 patients, respectively, for KCCQ score.

Baseline demographic and clinical characteristics of the study population are shown in Table 1. Participants had a mean (± SD) age of 54.9 (± 15.0) years, and most were male (n = 507, 77.4%). The median time since HF diagnosis was 5 months and the mean (± SD) LVEF was 30.9 (± 7.2)%. Most (> 75%) patients had a baseline HR of 80 bpm or higher, but mean (± SD) systolic and diastolic blood pressures were normal (116.3 [± 17.3] mmHg and 74.9 [± 12.3] mmHg, respectively). In 81.8% of cases (n = 536), HF was due to either dilated cardiomyopathy or coronary heart disease; collectively, hypertension, valvular heart disease, and myocarditis accounted for a minority (n = 81, 12.3%) of cases. Most patients had either NYHA class II or III HF (n = 511, 78.0%) and almost all were receiving treatment with a β-blocker (85.3%) and/or an aldosterone receptor antagonist (88.5%).

Table 1 Demographics and baseline clinical data of the 655 patients included in the full analysis set of the interim analysis of the POSITIVE study
Full size table

At baseline, most patients were receiving ivabradine at a dose of 2.5 mg (n = 363, 55.4%) or 5 mg (n = 288, 44.0%), and most patients received twice-daily administration (n = 620, 94.7%). Similarly, at M1, most patients received a dose of 2.5 mg (n = 211/509, 41.5%) or 5 mg (n = 262/509, 51.5%), administered twice daily (n = 488/509, 95.9%). Changes in study drug administration from the previous visit were as follows: 369 patients (68.8%) had no change in dose, 120 (22.4%) had a dose increase, 21 (3.9%) had a dose decrease, and 26 (4.9%) discontinued treatment. At M6, 2.5 mg (n = 131/354, 37.0%) and 5 mg (n = 192/354, 54.2%) continued to be the most commonly used doses, and most patients received twice-daily administration (n = 335/354, 94.6%). At this time point, changes in study drug administration from the previous visit were: 309 patients (73.0%) had no change of dose, 25 (5.9%) had a dose increase, 19 (4.5%) had a dose decrease, and 70 (16.5%) discontinued treatment.

3.2 Heart Rate

Mean HR (± SD) declined from 88.9 (± 12.4) bpm at baseline to 75.0 (± 12.7) bpm at M1 and 73.8 (± 12.8) bpm at M6, representing decreases of 13.2 (95% confidence interval [CI] 11.2–15.2) bpm between baseline and M1, and 14.5 (95% CI 11.8–17.2) bpm between baseline and M6 (Fig. 1a). The reductions in HR at both follow-up visits were statistically significant (p < 0.001).

Fig. 1
figure 1

a Mean heart rate; b New York Heart Association (NYHA) functional class; and c mean Kansas City Cardiomyopathy Questionnaire scores at baseline and at the two follow-up visits (Month 1 and Month 6). Whiskers in (a, c) signify standard deviation. * p < 0.001 versus baseline. bpm beats per minute

Full size image

3.3 Symptoms and Cardiac Function

NYHA class showed significant improvements at both follow-up visits compared with baseline, with all p-values <0.001 (Fig. 1b). Over the course of 6 months, the proportion of patients with NYHA class III or IV HF decreased, whereas the proportion of patients with NYHA class I or II HF increased. An improvement in NYHA class of at least one class relative to baseline was seen in 58.3% of subjects at M1 and 63.1% of subjects at M6.

Levels of BNP and NT-proBNP decreased during the study (Fig. 2), but the reductions from baseline were not statistically significant. LVEF tended to increase: median (IQR) LVEF was 31.0% (26.0–36.0%) at baseline, 35.0% (29.0–43.0%) at M1, and 41.0% (35.0–52.9%) at M6.

Fig. 2
figure 2

Median plasma levels of BNP and NT-proBNP at baseline, Month 1 and Month 6. Whiskers signify the interquartile range. BNP B-type natriuretic peptide, NT-proBNP N-terminal pro-BNP

Full size image

3.4 Quality of Life

Mean KCCQ clinical symptom and total scores of KCCQ were significantly higher at both M1 and M6 compared with baseline (p < 0.001 for all comparisons). Most patients had changes of ≥ 5 points in clinical symptom and total scores between baseline and both follow-up visits (65.5% at M1 and 71.4% at M6, and 65.3% at M1 and 73.9% at M6, respectively), indicating clinically meaningful improvements in QoL.

3.5 Safety and Tolerability

In total, 359 AEs occurred in 205 subjects (31.3%) (Table 2). All AEs were treatment-emergent, and 60 (reported by 44 subjects [6.7%]) were considered to be related to ivabradine. The most common drug-related AEs were phosphenes and bradycardia, which occurred in six patients each (0.9%). Eleven patients developed a total of 14 serious drug-related AEs, including atrial fibrillation (n = 3, 0.5%), HF (n = 2, 0.3%), hypotension (n = 2, 0.3%), sinus bradycardia (n = 1, 0.2%), ventricular extrasystole (n = 1, 0.2%), bradycardia (n = 1, 0.2%), myocardial infarction (n = 1, 0.2%), palpitations (n = 1, 0.2%), prolonged QT interval on ECG (n = 1, 0.2%), and atopic dermatitis (n = 1, 0.2%).

Table 2 Adverse events and treatment-emergent adverse events occurring in the full analysis set
Full size table

4 Discussion

We report interim data from the first large real-world study of ivabradine in Chinese patients with HFrEF. In our cohort of 655 patients, ivabradine significantly reduced mean HR by 13.2 (95% CI 11.2–15.2) bpm and 14.5 (95% CI 11. 8–17.2) bpm after 1 and 6 months of treatment, respectively. In comparison, placebo-corrected reductions in HR among ivabradine recipients in the SHIFT trial were 10.9 (95% CI 10.4–11.4) bpm for the overall study population (n = 3241) [13], and 12.1 (95% CI 8.9–15.3) bpm among the Chinese subpopulation (n = 225) [20].

We also found clinically important improvements in NYHA class over 6 months, with approximately 60% of patients experiencing an improvement of at least one NYHA class (e.g., class III to class II) at both follow-up visits. Concomitant reductions in HR and improvements in NYHA class were also found in the SHIFT trial [8] of ivabradine and in the real-world INTENSIFY study [15].

Treatment with ivabradine also improved QoL in our study, as indicated by increases in KCCQ scores over time. This finding is consistent with the results of previous prospective [15,16,17] and retrospective [21] real-world studies of ivabradine in patients with HF. These findings are important not only because improving symptoms and QoL is a major goal of HF management [12, 22], but also because such improvements can have prognostic significance [23,24,25,26]. For example, in the SHIFT trial [26], a low baseline KCCQ score was predictive of cardiovascular death or hospital admission in patients with NYHA class II–IV systolic HF [26].

We found trends towards reductions in blood BNP and NT-proBNP levels, and towards improvements in LVEF. Because of the observational nature of our study, monitoring reflected routine clinical practice in China, and the fact that current national guidelines on the management of HF do not recommend regular echocardiography or routine measurement of BNP or NT-proBNP levels [22]. Thus, many of our patients had missing values and this may explain why the trends we observed did not reach statistical significance.

Treatment-emergent AEs were broadly consistent with the known tolerability profile of ivabradine and there were no new or unexpected findings in relation to drug safety. Our findings therefore suggest that ivabradine is generally well tolerated in Chinese patients with HFrEF.

Ivabradine is an important addition to the available treatment options for HF. β-blockers, although effective in reducing HR, alter myocardial contractility and intracardiac conduction; this can preclude their use or prevent dose optimization in some patients. In contrast, ivabradine reduces HR without affecting conduction or contractility [8]. In addition to the clinical effects described above, ivabradine has been found to improve left ventricular remodeling [14], increase exercise tolerance [27], and reduce the risk of HF-related death [13].

The mechanism by which HR reduction improves outcomes in HF has been investigated. A key finding has been that, in HF, myocardial contractility is inversely proportional to HR [1]; thus, reducing HR in patients with HF should increase myocardial contractility, and this has been demonstrated with ivabradine [2]. In addition, the ivabradine-induced HR reduction increases the diastolic interval, which in turn, improves coronary perfusion [28] and results in greater ventricular filling during diastole, which increases stroke volume [29]. Together, these mechanisms reduce myocardial oxygen demand and explain improvements in cardiac output associated with ivabradine treatment [29]. These mechanisms are achieved without negative inotropic effects or a negative impact on conduction [30, 31]. Long-term, ivabradine treatment can delay or even reverse myocardial apoptosis and remodeling induced by energy deficiency [14], ultimately improving cardiac function and QoL.

Until now, the main source of clinical support for the use of ivabradine in Chinese patients with HF has been a subanalysis of 225 patients in the SHIFT trial [20]. As described earlier, the mean placebo-corrected reduction in HR in patients receiving ivabradine was 12.1 bpm at 28 days. However, the primary endpoint of the SHIFT trial was a composite of cardiovascular death or hospitalization for worsening HF; in the Chinese subpopulation, ivabradine was associated with a 44% reduction in the incidence of this outcome versus placebo (hazard ratio [HR] 0.56, 95% CI 0.33–0.97; p = 0.039), after a mean follow-up of 15.6 months. Because of the study design and type of patients included in the POSITIVE study, we did not investigate the effects of ivabradine on mortality or hospitalization rates.

The mean age of patients in our cohort (54 years) is younger than previously reported in Chinese registry studies of hospitalized patients with HF [32, 33]. For example, the mean age of patients hospitalized with HFrEF in the HERO registry was 66.6 years [32]. This likely reflects the underlying cause of HF in our population, since 45.6% had dilated cardiomyopathy and 36.2% had coronary artery disease, and patients with dilated cardiomyopathy tend to be 5–10 years younger than patients with HF caused by other etiologies [34]. In contrast, coronary artery disease was the leading cause of HF in the China PEACE study (affecting 60.6% of patients), in which the median age was 73 years [33]. In addition, approximately 25% of patients in our cohort were outpatients, whereas the previous registry studies were in hospitalized patients, which may also explain the difference in age between our cohort and previous reports. The mean age of our cohort may also reflect the type of patients who are prescribed ivabradine, since data show that HF patients who would meet the criteria for entry to the SHIFT study are significantly younger than patients who would not have been candidates for ivabradine in that study [35].

This study has limitations inherent in the nature of its design (open-label, observational), which may have resulted in bias towards overestimating the treatment effect. Other limitations are the absence of a placebo group and the fact that changes in background therapies might also lead to overestimation of the treatment effect. In addition, the results of observational studies can be undermined by missing data, because (in contrast to clinical trials) measurements of interest are not routinely or regularly performed in clinical practice. Lastly, our findings should be regarded as preliminary; the POSITIVE study is ongoing, with more than 1000 patients enrolled to date and a final analysis to be performed once enrolment and follow-up are complete.

5 Conclusion

The POSITIVE study is a multicenter, prospective, observational study of ivabradine in the treatment of Chinese patients with HF and is the largest study of ivabradine undertaken to date in a Chinese population. Interim data show that ivabradine treatment reduced HR over 6 months and was associated with statistically significant improvements in NYHA class and clinically meaningful changes in QoL scores. These benefits were accompanied by good general tolerability.