Sustained-release CARvedilol in patients with hypErtension and Heart failure with Preserved ejection fraction (CARE-preserved HF) is a multicenter, prospective, randomized, double blinded, phase 4 pilot study that assesses the efficacy of carvedilol-SR compared with placebo in patients with hypertension and HFpEF. The study flow chart is presented in Figure 1. Three tertiary university hospitals in South Korea participated in the study. Enrollment began in November 2021, is ongoing and is expected to be completed by early 2024. The trial design was registered at ClinicalTrial.gov (NCT05553314).

We have been enrolling patients with hypertension and HFpEF aged ≥20 years. Patients must have symptoms of HF and elevated natriuretic peptides at the screening visit: NT-proBNP ≥220 pg/ml [or B-type natriuretic peptide (BNP) ≥80 pg/ml] for patients with sinus rhythm (SR), or ≥660 pg/ml (or BNP ≥240 pg/ml) for patients with atrial fibrillation (AF). For the hypertension criteria: (1) patients' average systolic blood pressure (SBP) must be ≥140 mmHg or diastolic blood pressure ≥90 mmHg, obtained from BP measured three times on the reference arm in the sitting position at screening; (2) for those taking anti-hypertensive medications who are on stable antihypertensive dosage that has not been adjusted for 8 weeks, the SBP must be ≥110 mmHg. Patients with SBP <110 mmHg or resting HR <60 beats/min at the screening visit will be excluded. For the HFpEF criteria, we referred to the consensus recommendation of the Heart Failure Association of the European Society of Cardiology (31). LVEF should be ≥50% on echocardiography within 2 months before randomization. In addition, patients should have one or both of the following evidence of structural heart disease: (a) left atrial volume index ≥29 ml/m² in patients with SR or ≥34 ml/m² in patients with AF, (b) left ventricular mass index ≥115 g/m² in males or ≥95 g/m² in females. Patients should also have at least one indication of functional heart disease in terms of average E/e', septal and lateral e', peak tricuspid regurgitation velocity, and pulmonary artery systolic pressure. Patients will be excluded if they are taking any beta-blockers within 4 weeks prior to enrollment or if they have a contraindication to beta-blockers. Patients with severe chronic kidney disease (creatinine >2.4 mg/dl) or liver enzyme levels three times the normal upper limit will be excluded. Detailed inclusion and exclusion criteria are shown in Table 1.

Potential participants will be screened (visit 1) on an outpatient or inpatient basis (Figure 2). After a comprehensive interview, eligible participants will be asked to provide written informed consent to participate in the trial. A baseline survey of eligible participants covering demographics and cardiovascular comorbidities will be conducted by research nurses (Table 2). Then, using a web-based central randomization service (http://matrixmdr.com), eligible participants will then be randomly assigned to either the carvedilol-SR group or the placebo group in a 1:1 ratio (visit 2). The randomization block will be generated by an independent statistician unrelated to this study using the SAS randomization program.

After randomization, patients in the carvedilol-SR group will receive carvedilol-SR 8, 16, 32, or 64 mg once daily for 6 months. Carvedilol-SR (trade name: Dilatrend SR) was manufactured and distributed by Chong Kun Dang Pharmaceutical Corporation in South Korea. The carvedilol dose will be up-titrated at the discretion of the treating physician. The use of beta-blockers other than carvedilol will not be allowed during the study. To avoid drug-drug interactions that may cause confounding effects, drugs with potential pharmacodynamic and pharmacokinetic interference will be prohibited during clinical trials (Table 3).

Follow-up visits will be scheduled at 3 (visit 3) and 6 (visit 4) months after randomization (Figure 2 and Table 2). Each clinic visit comprises history-taking, physical examination, laboratory tests, and echocardiography. Drug compliance, adverse effects, and the use of additional concomitant drugs will also be checked. Unscheduled visits will be recorded.

The outcomes of this study are presented in Table 4. The primary efficacy outcome will be the time-averaged proportional change in NT-proBNP concentration from baseline through months 3 and 6. The secondary endpoints included the proportion of patients with an NT-proBNP decrease >20% from baseline through months 3 and 6, composite of all-cause mortality and all-cause hospitalization at 6 months. We will also assess changes in BP, HR, New York Heart Association (NYHA) classification, and quality of life at 6 months. Rate of BP control were under controlled will be compared between the two groups, which is defined as SBP <140 mmHg and diastolic BP <90 mmHg at 6 months. In addition, drug adherence and changes in echocardiographic parameters, including left atrial volume index, left ventricular mass index, E/e', e', pulmonary artery systolic pressure, and GLS will be assessed between the two groups. All adverse reactions and the incidence of the following adverse events of special interest during treatment will be assessed: symptomatic hypotension and bradycardia (investigator-reported).

Blood sampling tests at the screening visit will be conducted by the laboratories at each participating institute, which are certified by the Korean Association of Quality Assurance for Clinical Laboratory. Measurement of NT-proBNP by an electro-chemiluminescence immunoassay method using cobas® 8,000 (Roche Diagnostics, Mannheim, Germany) will be performed in a central laboratory at randomization and at 6 months (visit 4). Quality of life will be assessed using a visual analog scale ranging from 0 to 10, with 0 being the worst and 10 being best. Drug adherence will be assessed by “pill count” measurements. The patients will bring the remaining tables to each scheduled visit, and trained and certified researchers will count the number of returned drugs and calculate drug adherence as follows:Drug adherence =number of pills dispensed-number of pills returnednumber of days between dispensing date and follow-up dateBaseline echocardiographic data collected within 2 months before randomization will be used. All images will be obtained using a standard ultrasound system with 2.5 MHz probes manufactured by GE, Phillips, and Siemens. Standard techniques will be used to obtain the M-mode, two-dimensional, and Doppler measurements. The LV dimensions, LVEF, and other echocardiographic parameters will be obtained according to the guidelines of American Society of Echocardiography and the European Association of Cardiovascular Imaging (32). Echocardiographic parameters at each institution will be archived in the Digital Imaging and Communications in Medicine format. GLS will be evaluated by a central echocardiography laboratory at the Samsung Medical Center, Seoul, Korea. GLS will be assessed using two-dimensional speckle-tracking echocardiography (2D-STE) on index echocardiography (33). The 2D-STE data will be analyzed using TomTec software (Image Arena 4.6, Munich, Germany) for deformation analyses (2-dimensional cardiac performance analysis) (27). For deformation analysis, endocardial borders are traced on the end-systolic frame in three apical views (4-, 2-, and 3-chamber), with end-systole defined by the QRS complex or as the smallest LV volume during the cardiac cycle. The software tracks speckles along the endocardial border and myocardium throughout the cardiac cycle. The peak longitudinal strain is computed automatically, generating regional data from six segments (anterior, anteroseptal, anterolateral, inferior, inferoseptal, and inferolateral) and an average value for each view. The GLS is determined as the average peak longitudinal strain of 18 LV segments from standard apical views. For patients with SR, analyses will be performed on a single cardiac cycle; for patients with AF, the strain values will be calculated as the average of three cardiac cycles. All strain measurements will be performed by strain specialists who are blinded to other data of each patient. All strain analyses will be performed by a single investigator. In our previous study, intra-observer variability in GLS was assessed in 30 randomly selected patients (34). The coefficient of variation has been identified as 5.8% for the GLS, and the intraclass correlation coefficient was 0.95 for GLS [95% confidence interval (CI), 0.91–0.98].

To date, no previous study has evaluated the effects of carvedilol-SR on HFpEF. Due to the absence of prior research and the pioneering nature of our intervention, a sample size calculation was not conducted at this preliminary stage. As this was a pilot study to explore the benefit of carvedilol-SR for HFpEF, we decided to enroll a total of 100 participants (50 in the carvedilol-SR group and 50 in the control group) across the three institutes, considering the duration of the study and the number of participating hospitals.

Categorical variables will be reported as frequencies (percentages), and continuous variables will be expressed as means ± standard deviations or medians with interquartile ranges. Categorical variables will be compared using Pearson's chi-square test or Fisher's exact test, and continuous variables will be compared using Student's t-test or the Mann–Whitney U-test.

Primary analysis of the proportional change in NT-proBNP concentration from baseline on a logarithmic scale will be performed using an analysis of covariance model with an adjustment for the baseline value. This will be calculated as the average of the geometric means of NT-proBNP at months 3 and 6 divided by the geometric mean of NT-proBNP at baseline in a natural logarithmic scale (i.e., ratio of geometric means) and summarized as the difference between the treatment groups in the ratio of the geometric means. Cumulative clinical-event rates will be calculated according to the Kaplan–Meier method, and the differences in clinical outcomes between the two treatment groups will be assessed using the log-rank test. Hazard ratios and associated 95% CIs will be calculated using a Cox proportional-hazards model. The incidence of adverse events will be calculated as relative risks with associated 95% CIs.

Missing values due to unevaluable samples or early study discontinuation will not be imputed. Patients with missing baseline NT-proBNP and/or missing 3 and 6 months will not be included in the primary analysis. The data will be primarily analyzed using intention-to-treat analysis, including all randomized patients. We will also perform a per-protocol analysis, which include all randomized patients who receive at least one dose of the study drug during the double-blind period and have no major protocol deviations. Safety outcomes will be analyzed in the safety set of patients who received the trial drug at least once.

Primarily, we will assess the efficacy of carvedilol-SR on the reduction of NT-proBNP in HFpEF compared with a placebo. In addition, we will evaluate the differential effects of carvedilol-SR on primary outcomes according to prespecified variables, including baseline GLS, and NT-proBNP, LVEF. We planned to use the cutoff value of 14% GLS, taking into account a previous study (28), and also planned to use the median cutoff value of the GLS.

All tests will be two-tailed, and a P-value <0.05 will be considered statistically significant. Statistical analyses will be performed using R version 4.2.3 (The R Foundation for Statistical Computing, Vienna, Austria).

This study had several limitations. First, it will only include East Asian patients, raising concerns about the extrapolation of these results to other ethnicities. Second, the number of participants to be enrolled in this study is not large, and it has a pilot study perspective. Third, the carvedilol dose will be up-titrated at the discretion of the treating physician and not according to the standard protocol in this trial, which could be a confounding factor. Fourth, the primary endpoint, change in NT-proBNP levels is a surrogate marker and may be influenced by confounding factors, such as the usage or dose of diuretics, which may be unequally controlled between the two groups. A high body mass index may also affect the level of this biomarker. Fifth, there will be no objective measure of functional status in terms of the 6-minute walk test or maximal oxygen uptake from the cardiopulmonary exercise testing. Despite these limitations, a major strength is that this study is a multicenter, randomized, double-blinded controlled trial with a novel design to evaluate the efficacy of carvedilol-SR in patients with hypertension and HFpEF and to assess the differential effects of these drugs according to GLS.