Effect of empagliflozin with or without carvedilol on tricuspid regurgitation and right ventricular remodeling: PROVE trial Authors: Ga Yun Kim, So-Min Lim, Dong Hyun Yang, Jong Eun Lee, Sahmin Lee, Sung-Ji Park, In-Chang Hwang, Yeonyee Elizabeth Yoon, Chi Young Shim, Hyung-Kwan Kim, Sung-Cheol Yun, Jong-Min Song, Duk-Hyun Kang Published: 2026 ABSTRACT AbstractBackgroundFunctional tricuspid regurgitation (TR) is associated with adverse outcomes. As TR, right ventricular (RV) remodeling, and worsening TR form a vicious cycle, therapies that reverse RV remodeling and reduce TR are needed. We examined whether sustained-release carvedilol (carvedilol SR), empagliflozin, or their combination reduces RV volumes and TR severity in patients with significant functional TR.MethodsIn this randomized trial with a 2 × 2 factorial design, 56 patients were assigned to carvedilol SR plus empagliflozin, carvedilol SR, empagliflozin, or placebo. The primary end point was the change in RV end-systolic volume index from baseline to 12 months. Secondary end points included changes in RV end-diastolic volume index, RV ejection fraction, vena contracta width (VCW), effective regurgitant orifice area (EROA), and regurgitant volume. Cardiac magnetic resonance imaging assessed RV volumes, and echocardiography evaluated TR severity.ResultsBaseline characteristics were well balanced between treatment groups (mean age 71 ± 7 years; men 55%; diabetes 14%; atrial fibrillation 89%). Empagliflozin did not change RV volumes or RV ejection fraction. However, empagliflozin reduced VCW (1.45 mm; 95% confidence interval [CI], 0.69–2.20), EROA (0.09 cm2; 95% CI, 0.02–0.17), and regurgitant volume (10.7 mL; 95% CI, 4.2–17.2) compared with placebo. Carvedilol SR showed no significant effect on primary or secondary end points. Compared with placebo, carvedilol SR plus empagliflozin reduced VCW (− 0.78 ± 1.15 mm vs. 0.56 ± 1.45 mm, P = 0.03) but did not change other outcomes.ConclusionsCarvedilol SR, empagliflozin, or their combination showed no significant reduction in RV volumes, but empagliflozin significantly reduced TR severity in patients with functional TR.Trial registrationClinicalTrials.gov identifier: NCT04345796 (PROVE trial).Keywords: Tricuspid regurgitation, Multimodality imaging, Doppler echocardiography, Sodium-glucose cotransporter 2 inhibitor Background The presence of significant tricuspid regurgitation (TR) is associated with substantial morbidity and poor long-term survival [1–3]. In patients with symptomatic, severe primary TR, tricuspid valve surgery can be beneficial, but management of secondary TR accounting for most cases of severe TR remains controversial [3–5]. Secondary, functional TR is commonly observed in patients with atrial fibrillation, heart failure (HF), and left-sided valve disease, and management of the underlying etiologies may decrease the severity of functional TR [5]. However, guideline-directed medical therapy (GDMT) for HF often fails to improve TR, and diuretics are provided to relieve systemic congestive symptoms associated with TR [5–7]. Functional TR may progress over time, because TR causes dilatation of the right ventricle (RV), which can subsequently result in further enlargement of tricuspid annulus and tethering of the leaflets, finally leading to deterioration of TR [8, 9]. As a vicious cycle of TR, RV remodeling, and worsening TR results in RV failure, which is closely related to poor prognosis, therapies that induce reversal of RV remodeling and consequently reduce TR may improve clinical outcomes. However, there have been no proven medical therapies for adverse remodeling of the RV and secondary TR [5]. Sodium-glucose cotransporter 2 (SGLT2) inhibitors reduce the risk of cardiovascular death and hospitalization for HF [10–13], and are included in the GDMT for HF [6]. SGLT2 inhibitors may also be a therapeutic option for RV failure because of their diuretic and nephroprotective effects, but their effects on RV remodeling and functional TR are unknown. Activation of the sympathetic nervous system contributes to progression of right- and left-sided HF via its effects on adverse ventricular remodeling [14]. While beta-blockers are a standard medical therapy for left ventricle (LV) failure [6], their use in managing RV failure is not well established. Therefore, the PROVE trial (Pharmacologic Reduction of Right Ventricular Enlargement) was designed to examine whether treatment with a SGLT2 inhibitor; empagliflozin, the beta-blocker; sustained-release carvedilol (carvedilol SR), or their combination would reduce RV volumes, compared with placebo. Furthermore, we also sought to evaluate the therapeutic efficacy of empagliflozin, carvedilol SR, or their combination on functional TR. Methods Study design PROVE was a prospective, multicenter, randomized, investigator-initiated trial with a 2-by-2 factorial design. The trial was conducted at five centers in Korea. In the first randomized comparison, we examined the effect of treatment with carvedilol SR at a daily dose of 16 mg, as compared with placebo, on the changes in RV volumes (primary outcome) and severity of TR (secondary outcome). In the second randomized comparison, we compared primary and secondary outcomes of patients assigned to GDMT plus empagliflozin 10 mg per day with those assigned to GDMT alone. We also assessed the effects of carvedilol SR plus empagliflozin, as compared with placebo. The study protocol was designed by the principal investigator and approved by the institutional review board at each participating center. The first draft of the manuscript was prepared by the first author, and was reviewed and edited by all the authors. All the authors made the decision to submit the manuscript for publication and vouch for the fidelity of this report to the protocol and for the accuracy and completeness of the data and the analyses. Patient selection Patients were eligible for enrollment if they were at least 20 years of age; had stable HF with New York Heart Association (NYHA) class II or III symptoms, with or without diabetes; had an LV ejection fraction (LVEF) ≥ 50%; and had significant functional TR, defined as a vena contracta width ≥ 0.7 cm or central jet area > 10 cm2, persisting for > 6 months despite GDMT. Patients were required to receive a stable, optimized GDMT for at least 4 weeks before enrollment. Exclusion criteria included symptomatic hypotension, a systolic blood pressure < 90 mmHg, bradycardia with heart rate < 50 beats/min or 2nd or 3rd degree atrioventricular block, LVEF < 50%, a glycated hemoglobin level > 10.5%, an estimated glomerular filtration rate < 30 mL/min per 1.73 m2, or a history of ketoacidosis. Patients were also excluded from the trial if they had any evidence of structural tricuspid valve disease; previous tricuspid valve intervention; significant left-sided valve disease, severe pulmonary hypertension; NYHA IV symptoms; previous hospitalization within 6 weeks; previous history of acute coronary syndrome, stroke, percutaneous coronary intervention, or cardiovascular surgery within 3 months; an indication for cardiac resynchronization therapy; a plan of coronary revascularization, tricuspid valve intervention, or heart transplantation during the trial. All the participants provided written informed consent. Study procedures Eligible patients entered a run-in phase lasting 4 weeks, during which they received placebo alone. Patients who had at least 80% adherence to placebo were randomly assigned to receive GDMT for HF alone or GDMT plus 10 mg of empagliflozin, carvedilol SR or matching placebo, with the use of a computerized randomization system involving concealed study-group assignments. The randomization was stratified according to presence or absence of diabetes and cardiac rhythm. The interactive web response system assigned a randomization number to each patient, which linked the patient to a treatment group, to keep the study information confidential and specified a unique medication number for the study drug to be dispensed. Carvedilol SR and the matching placebo were identical in appearance. All other treatments, including GDMT, were continued. Patients were treated for 12 months and evaluated every 4 weeks during the first 2 months of therapy and every 2 to 6 months thereafter. Cardiac magnetic resonance (CMR) imaging and echocardiographic evaluation were performed at randomization and at the 12-month follow-up visit. CMR was conducted using the latest 3.0 T magnetic resonance imaging scanner equipped with phased-array cardiac coils at participating centers, all of which had been approved by a central CMR core laboratory for their equipment and compliance with the specified acquisition protocols. CMR data collected at each site were anonymized and electronically transferred to a central server (AiCROsystem, Asan Image Metrics) for image storage and unbiased, independent image analysis [15]. Cine CMR images were analyzed using a standardized technique with commercially available software (CMR 42, Circle Cardiovascular Imaging) [16]. The short-axis cine images, spanning from the base to the apex of the RV, were measured by an investigator (DHY) who was blinded to patient data, treatment allocation, study sequence, and any previous measurements. Experienced sonographers who were certified by the echocardiography core laboratory and dedicated to this trial performed a standard two-dimensional and Doppler echocardiographic examination on all patients using an iE33 echocardiography system (Philips). Echocardiographic studies were stored digitally in Digital Imaging and Communications in Medicine format and transferred to the echocardiography core laboratory, where measurements were performed using an offline digital review workstation by an investigator (DHK) who was masked to all the information related to the trial. The end-systolic volume, end-diastolic volume, and EF of the LV and end-systolic volume of the left atrium were calculated with the biplane Simpson method [17]. For vena contracta measurement, the ultrasound beam was oriented as perpendicular to the regurgitant flow as possible, and the width of the narrowest portion of TR jet flow at or immediately downstream of the regurgitant orifice was measured in a zoomed view. Vena contracta width was measured on RV modified apical 4-chamber and RV inflow views, and the measurements were averaged [18, 19]. Effective regurgitant orifice area (EROA) was determined by dividing the regurgitant flow rate, calculated as 2πr2 × aliasing velocity, where r is the proximal isovelocity surface area (PISA) radius, by peak TR velocity [19]. The regurgitant volume was estimated as EROA multiplied by the velocity time integral of the TR jet. Outcomes The primary end point was the change in RV end-systolic volume index from baseline to the 12-month follow-up. Secondary end points included changes in RV end-diastolic volume index, RV ejection fraction (RVEF), vena contracta width, EROA of TR, and regurgitant volume. Statistical analysis Based on our previous study [20], we assumed a common baseline mean RV end-systolic volume index of 88.5 mL/m2, a common standard deviation of 30.1 mL/m2. Given these assumptions, we calculated that a sample size of 164 patients randomly assigned to two groups would provide 80% power to detect a difference of 15% in the RV end-systolic volume index between the carvedilol SR and placebo groups, using a two-sided t-test with an alpha level of 0.05. The study plan was to randomize 180 patients, allowing for a drop-out rate of 10%. Because the COVID-19 pandemic seriously affected enrollment of study patients, resulting in a lower rate of patient enrollment than expected, the sponsor had a logistic problem in providing study drugs beyond the planned timeline. On the basis of a limited supply of study drug, the executive committee decided to stop enrollment in December 2023 without knowledge of any trial data and the actual number of study patients was 56. The primary analysis for each treatment comparison was based on all participants who had undergone randomization. If no interactions were observed between the main effects of carvedilol SR and empagliflozin, each individual comparison was to be performed. Analyses followed the intention-to-treat principle, with all randomized patients analyzed in their originally assigned groups. Baseline clinical and echocardiographic characteristics were compared in the two treatment groups using Student’s t-test or the Mann–Whitney U test for continuous variables and the chi-square test or Fisher’s exact test for categorical variables as appropriate. The null hypothesis was that there would be no difference between the treatment groups regarding the change in RV end-systolic volume index from baseline to 12-month follow-up, and an intention-to-treat analysis tested this hypothesis. For the analyses of primary and secondary end points, we used the t-test for differences between groups as described in the protocol. Estimated between-group differences are presented with 95% confidence interval (CI). All reported P-values were two-sided, and P < 0.05 was considered statistically significant. SAS ver. 9.1 (SAS Institute) and R ver. 3.3.1 (R Foundation for Statistical Computing) were used for statistical analyses. Results Patient characteristics Between January 2021 and July 2023, a total of 56 patients completed the run-in phase; 13 patients were randomly assigned to receive carvedilol SR plus empagliflozin, 14 to receive carvedilol SR, 14 to receive empagliflozin, and 15 to receive placebo. CMR and echocardiography were performed at a median interval of 28 days (interquartile range, 0 to 37.5 days) across the baseline and 12-month assessments, with both studies performed on the same day in 38% of assessments. The treatment groups were generally well balanced with regard to baseline clinical, CMR, and echocardiographic characteristics, except for age and RV end-systolic volume index (Table 1). The mean (± SD) age of the patients was 70.9 ± 6.8 years and 31 (55%) were men, 8 (14%) had diabetes, 54 (96%) were in NYHA functional class II and 2 (4%) in NYHA class III, and atrial fibrillation was present in 50 (89%). The mean LVEF was 60 ± 7%, the mean RV end-systolic volume index was 59 ± 22 mL/m2, the mean RV end-diastolic volume index was 113 ± 35 mL/m2, and the mean RVEF was 48 ± 9%. The mean vena contracta width was 8.6 ± 2.2 mm, the mean EROA of TR was 0.52 ± 0.25 cm2, and the mean regurgitant volume was 44 ± 19 mL. Most patients received GDMT, and diuretics were used in 39 (70%) patients and beta-blockers in 30 (54%). Table 1Baseline characteristics of the patients according to treatment groupCharacteristicCarvedilol SR plus empagliflozin(n = 13)Carvedilol SR alone(n = 14)Empagliflozin alone(n = 14)Placebo alone(n = 15)P-valueAge (yr)75 ± 572 ± 670 ± 768 ± 70.025Male sex5 (38.5)9 (64.3)6 (42.9)11 (73.3)0.187BMI (kg/m2)24.2 ± 3.324.4 ± 3.225.4 ± 2.725.5 ± 2.60.397Hypertension9 (69.2)9 (64.3)8 (57.1)9 (60.0)0.946Diabetes2 (15.4)1 (7.1)2 (14.3)3 (20.0)0.918Atrial fibrillation12 (92.3)14 (100)11 (78.6)13 (86.7)0.385Chronic kidney disease3 (23.1)3 (21.4)4 (28.6)1 (6.7)0.490Coronary artery disease2 (15.4)1 (7.1)2 (14.3)5 (33.3)0.337NYHA functional class0.051II11 (84.6)14 (100)14 (100)15 (100)III2 (15.4)0 (0)0 (0)0 (0)Heart failure medicationARNI1 (7.7)1 (7.1)1 (7.1)1 (6.7) > 0.999ACE inhibitor0 (0)1 (7.1)0 (0)0 (0)0.732ARB5 (38.5)4 (28.6)6 (42.9)10 (66.7)0.222Beta-blocker8 (61.5)7 (50.0)6 (42.9)9 (60.0)0.762Diuretic12 (92.3)8 (57.1)9 (64.3)10 (66.7)0.194Aldosterone antagonist3 (23.1)3 (21.4)2 (14.3)1 (6.7)0.609Systolic blood pressure (mmHg)129 ± 12119 ± 14128 ± 15127 ± 150.357Diastolic blood pressure (mmHg)76 ± 773 ± 777 ± 1373 ± 110.634Heart rate (beats/min)75 ± 1178 ± 983 ± 1273 ± 110.111Hemoglobin (g/dL)13.3 ± 1.513.3 ± 1.613.2 ± 1.614.4 ± 1.60.229Albumin (g/dL)4.0 ± 0.34.0 ± 0.54.2 ± 0.34.0 ± 0.30.125Creatinine (mg/dL)0.91 ± 0.260.92 ± 0.220.95 ± 0.190.96 ± 0.190.862Glucose (mg/dL)123 ± 53105 ± 23118 ± 22113 ± 330.270HbA1c (%)6.3 ± 1.45.7 ± 0.66.1 ± 0.55.9 ± 0.60.109Cholesterol (mg/dL)160 ± 26158 ± 36156 ± 31154 ± 360.993LDL cholesterol (mg/dL)83 ± 30100 ± 2177 ± 2973 ± 250.428NT-proBNP (pg/mL)1012 ± 4191289 ± 984638 ± 413897 ± 7060.075Echocardiographic measureLeft heart structure and functionLV end-diastolic dimension (mm)48.0 ± 6.450.5 ± 3.645.6 ± 4.850.8 ± 6.40.042LV end-systolic dimension (mm)31.2 ± 7.832.9 ± 5.029.4 ± 4.633.9 ± 5.30.096LVEF (%)59.9 ± 8.159.5 ± 7.158.7 ± 6.360.5 ± 5.30.939Left atrial volume index (mL/m2)71 ± 3264 ± 1554 ± 1676 ± 380.401E/e´ ratio10.9 ± 2.712.1 ± 8.39.6 ± 3.013.0 ± 7.00.290Right heart structure and functionVena contracta width (mm)9.20 ± 1.857.67 ± 2.079.29 ± 2.728.47 ± 2.060.149EROA (cm2)0.57 ± 0.220.48 ± 0.220.55 ± 0.310.51 ± 0.270.149Regurgitant volume (mL)46.9 ± 19.340.6 ± 19.444.6 ± 23.542.3 ± 16.70.810TAPSE (mm)19.6 ± 3.419.1 ± 4.219.4 ± 2.619.3 ± 3.80.998TR peak velocity (m/sec)2.7 ± 0.22.8 ± 0.32.7 ± 0.32.8 ± 0.30.554Cardiac magnetic resonance measureLV end-diastolic volume index (mL/m2)75.3 ± 12.775.6 ± 20.278.3 ± 29.481.3 ± 21.70.890LV end-systolic volume index (mL/m2)32.4 ± 10.534.7 ± 12.436.9 ± 27.939.5 ± 12.10.354RV end-diastolic volume index (mL/m2)104.1 ± 21.2103.5 ± 41.0112.9 ± 31.5130.5 ± 36.50.183RV end-systolic volume index (mL/m2)52.4 ± 14.350.1 ± 22.756.0 ± 15.975.4 ± 25.90.010RVEF (%)49.8 ± 7.651.0 ± 7.250.1 ± 7.342.3 ± 9.80.089Values are presented as number (%) or mean ± standard deviationACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor-neprilysin inhibitor; BMI, body mass index; carvedilol SR, sustained-release carvedilol; E/e’ ratio, the ratio of early diastolic mitral inflow velocity to early diastolic mitral annular tissue velocity at the septal annulus; EROA, effective regurgitant orifice area; HbA1c, glycated hemoglobin; LDL, low-density lipoprotein; LV, left ventricle; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal pro-B-type natriuretic peptide; NYHA, New York Heart Association; RV, right ventricle; RVEF, right ventricular ejection fraction; TAPSE, tricuspid annular plane systolic excursion; TR, tricuspid regurgitation Baseline characteristics of the patients according to treatment group Values are presented as number (%) or mean ± standard deviation ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; ARNI, angiotensin receptor-neprilysin inhibitor; BMI, body mass index; carvedilol SR, sustained-release carvedilol; E/e’ ratio, the ratio of early diastolic mitral inflow velocity to early diastolic mitral annular tissue velocity at the septal annulus; EROA, effective regurgitant orifice area; HbA1c, glycated hemoglobin; LDL, low-density lipoprotein; LV, left ventricle; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal pro-B-type natriuretic peptide; NYHA, New York Heart Association; RV, right ventricle; RVEF, right ventricular ejection fraction; TAPSE, tricuspid annular plane systolic excursion; TR, tricuspid regurgitation Follow-up Among the 13 patients randomly assigned to carvedilol SR plus empagliflozin, 3 patients withdrew from the trial (1 had a serious adverse event of bradycardia) and 10 completed the trial. In the carvedilol SR group (n = 14), 4 patients withdrew from the trial (1 had a serious adverse event of hyperkalemia) and 10 completed the trial. In the empagliflozin group (n = 14), 3 patients withdrew from the trial and 11 completed the trial. Of the 15 patients randomly assigned to placebo, 2 patients withdrew from the trial and 13 completed the trial. No patient underwent cardiac surgery or percutaneous intervention during the trial period. Although the original analysis plan specified an intention-to-treat analysis, the primary analysis included 44 patients who completed the study. Outcomes for carvedilol SR as compared with placebo After 12 months of treatment, the primary end point of change in RV end-systolic volume index was not significantly different between the carvedilol SR and placebo groups (–0.55 ± 9.80 mL/m2 vs. –4.75 ± 17.47 mL/m2, P = 0.323) (Table 2). The change in RV end-diastolic volume index (–0.65 ± 22.90 mL/m2 vs. –1.30 ± 27.94 mL/m2, P = 0.935) and RVEF (1.14 ± 9.59% vs. 2.62 ± 10.03%, P = 0.624) were similar between the treatment groups. Changes in vena contracta width of TR, EROA, and regurgitant volume also did not differ significantly between the treatment groups. Table 2Outcomes of carvedilol SR versus placeboOutcomeBaselineFollow-upChangeCarvedilol SR(n = 20)Placebo(n = 24)P-valueCarvedilol SR(n = 20)Placebo(n = 24)P-valueCarvedilol SR(n = 20)Placebo(n = 24)Difference (95% CI)P-valuePrimary end pointRV end-systolic volume index (mL/m2)50.5 ± 13.769.2 ± 24.00.00249.9 ± 16.664.5 ± 22.60.021–0.55 ± 9.80–4.75 ± 17.474.19 (–4.67 to 13.06)0.323Secondary end pointRV end-diastolic volume index (mL/m2)104.2 ± 25.8127.5 ± 35.20.018103.5 ± 28.9126.2 ± 42.90.051–0.65 ± 22.9–1.30 ± 27.940.64 (–15.12 to 16.41)0.935RVEF (%)50.9 ± 8.145.7 ± 9.80.06552.1 ± 9.048.4 ± 8.50.1671.14 ± 9.592.62 ± 10.03–1.47 (–7.48 to 4.53)0.624TR VC width (mm)8.08 ± 1.729.23 ± 2.440.0838.04 ± 1.859.10 ± 2.790.152–0.04 ± 1.25–0.12 ± 1.570.08 (–0.80 to 0.97)0.853TR EROA (cm2)0.52 ± 0.250.58 ± 0.290.5130.52 ± 0.270.57 ± 0.290.583–0.004 ± 0.14–0.01 ± 0.110.01 (–0.07 to 0.09)0.859TR regurgitant volume (mL)44.1 ± 21.647.1 ± 19.70.62544.4 ± 22.446.1 ± 20.90.8000.35 ± 14.16–1.00 ± 9.441.35 (–5.98 to 8.68)0.712Values are presented as mean ± standard deviationCarvedilol SR, sustained-release carvedilol; CI, confidence interval; EROA, effective regurgitant orifice area; RV, right ventricle; RVEF, right ventricular ejection fraction; TR, tricuspid regurgitation; VC, vena contracta Outcomes of carvedilol SR versus placebo Values are presented as mean ± standard deviation Carvedilol SR, sustained-release carvedilol; CI, confidence interval; EROA, effective regurgitant orifice area; RV, right ventricle; RVEF, right ventricular ejection fraction; TR, tricuspid regurgitation; VC, vena contracta Outcomes for empagliflozin as compared with placebo The primary end point of change in RV end-systolic volume index did not differ significantly between the empagliflozin and placebo groups (–1.97 ± 16.20 mL/m2 vs. –3.64 ± 13.07 mL/m2, P = 0.708) (Table 3). There was also no significant difference in the change in RV end-diastolic volume index (–2.72 ± 29.09 mL/m2 vs. 0.56 ± 22.28 mL/m2, P = 0.675) and RVEF (0.35 ± 9.15% vs. 3.41 ± 10.24%, P = 0.304) between the treatment groups. Regarding secondary end points of TR severity, the changes in vena contracta width (–0.82 ± 1.26 mm vs. 0.62 ± 1.20 mm, P < 0.001), EROA (–0.06 ± 0.15 cm2 vs. 0.04 ± 0.07 cm2, P = 0.013), and regurgitant volume (–5.86 ± 12.82 mL vs. 4.86 ± 7.77 mL, P = 0.002) were significantly different (Fig. 1). Compared with placebo, empagliflozin significantly reduced vena contracta width by 1.45 mm (95% CI, 0.69–2.20), EROA of TR by 0.09 cm2 (95% CI, 0.02–0.17), and regurgitant volume by 10.72 mL (95% CI, 4.23–17.22). Table 3Outcomes of empagliflozin versus placeboOutcomeBaselineFollow-upChangeEmpagliflozin(n = 21)Placebo(n = 23)P-valueEmpagliflozin(n = 21)Placebo(n = 23)P-valueEmpagliflozin(n = 21)Placebo(n = 23)Difference (95% CI)P-valuePrimary end pointRV end-systolic volume index (mL/m2)56.8 ± 14.864.3 ± 26.70.25454.8 ± 20.660.6 ± 21.90.371–1.97 ± 16.20–3.64 ± 13.071.67 (–7.25 to 10.59)0.708Secondary end pointRV end-diastolic volume index (mL/m2)113.7 ± 27.1119.8 ± 38.10.546111.0 ± 42.3120.4 ± 35.10.425–2.72 ± 29.090.56 ± 22.28–3.28 (–18.97 to 12.40)0.675RVEF (%)49.7 ± 7.546.6 ± 10.60.27650.1 ± 8.150.0 ± 9.60.9910.35 ± 9.153.41 ± 10.24–3.06 (–8.99 to 2.87)0.304TR VC width (mm)9.47 ± 2.238.00 ± 1.950.0268.64 ± 2.428.57 ± 2.490.926–0.82 ± 1.260.62 ± 1.20–1.45 (–2.20 to –0.69) < 0.001TR EROA (cm2)0.59 ± 0.270.51 ± 0.270.3500.54 ± 0.300.55 ± 0.270.858–0.06 ± 0.150.04 ± 0.07–0.09 (–0.17 to –0.02)0.013TR regurgitant volume (mL)48.7 ± 21.543.0 ± 19.50.36042.9 ± 21.747.6 ± 21.30.470–5.86 ± 12.824.86 ± 7.77–10.72 (–17.22 to –4.23)0.002Values are presented as mean ± standard deviationCI, confidence interval; EROA, effective regurgitant orifice area; RV, right ventricle; RVEF, right ventricular ejection fraction; TR, tricuspid regurgitation; VC, vena contracta Outcomes of empagliflozin versus placebo Values are presented as mean ± standard deviation CI, confidence interval; EROA, effective regurgitant orifice area; RV, right ventricle; RVEF, right ventricular ejection fraction; TR, tricuspid regurgitation; VC, vena contracta Baseline echocardiography shows tethering of tricuspid valve with severe functional tricuspid regurgitation (TR). Vena contracta width (VCW) was measured, effective regurgitant orifice area (EROA) and regurgitant volume were calculated with proximal isovelocity surface area (PISA) radius, peak velocity and velocity time integral of the TR jet (A). Tethering of tricuspid valve was improved and VCW, EROA, and regurgitant volume were significantly decreased after treatment with empagliflozin for 12 months (B). EROA, effective regurgitant orifice area; PISA, proximal isovelocity surface area; RVol, regurgitant volume; TV, tricuspid valve; VCW, vena contracta width Outcomes for carvedilol SR plus empagliflozin as compared with placebo Between the carvedilol SR plus empagliflozin and placebo groups, RV end-systolic volume index (–2.80 ± 10.09 mL/m2 vs. –7.74 ± 14.28 mL/m2, P = 0.364), RV end-diastolic volume index (–4.33 ± 20.90 mL/m2 vs. –1.33 ± 20.45 mL/m2, P = 0.733), and RVEF (1.36 ± 10.50% vs. 5.31 ± 10.96%, P = 0.392) did not differ significantly (Table 4). There was also no significant difference in the change in EROA (–0.06 ± 0.17 cm2 vs. 0.02 ± 0.07 cm2, P = 0.129) and regurgitant volume (–6.00 ± 16.24 mL vs. 3.33 ± 7.23 mL, P = 0.118) between the treatment groups, but carvedilol SR plus empagliflozin significantly reduced the vena contracta width by 1.34 mm (95% CI, 0.16–2.52), compared with placebo (P = 0.029). Table 4Outcomes of carvedilol SR plus empagliflozin versus placeboOutcomeBaselineFollow-upChangeCarvedilol SR plus Empagliflozin(n = 10)Placebo(n = 13)P-valueCarvedilol SR plus Empagliflozin(n = 10)Placebo(n = 13)P-valueCarvedilol SR plus Empagliflozin(n = 10)Placebo(n = 13)Difference (95% CI)P-valuePrimary end pointRV end-systolic volume index (mL/m2)54.1 ± 14.377.7 ± 27.10.02151.3 ± 18.269.9 ± 21.80.041–2.80 ± 10.09–7.74 ± 14.284.94 (–6.13 to 16.01)0.364Secondary end pointRV end-diastolic volume index (mL/m2)106.4 ± 20.4133.6 ± 38.20.055102.1 ± 29.4132.2 ± 35.20.040–4.33 ± 20.90–1.33 ± 20.45–3.00 (–21.06 to 15.06)0.733RVEF (%)49.1 ± 8.541.9 ± 10.50.09250.5 ± 8.547.2 ± 8.80.3841.36 ± 10.505.31 ± 10.96–3.95 (–13.37 to 5.46)0.392TR VC width (mm)8.93 ± 1.528.61 ± 2.090.6868.15 ± 2.149.12 ± 3.000.404–0.78 ± 1.150.56 ± 1.45–1.34 (–2.52 to –0.16)0.029TR EROA (cm2)0.55 ± 0.250.53 ± 0.290.8570.49 ± 0.300.56 ± 0.290.609–0.06 ± 0.170.02 ± 0.07–0.09 (–0.20 to 0.03)0.129TR regurgitant volume (mL)46.1 ± 21.543.8 ± 17.60.77740.1 ± 21.946.8 ± 20.40.470–6.00 ± 16.243.33 ± 7.23–9.33 (–20.18 to 1.51)0.118Values are presented as mean ± standard deviationCarvedilol SR, sustained-release carvedilol; CI, confidence interval; EROA, effective regurgitant orifice area; RV, right ventricle; RVEF, right ventricular ejection fraction; TR, tricuspid regurgitation; VC, vena contracta Outcomes of carvedilol SR plus empagliflozin versus placebo Values are presented as mean ± standard deviation Carvedilol SR, sustained-release carvedilol; CI, confidence interval; EROA, effective regurgitant orifice area; RV, right ventricle; RVEF, right ventricular ejection fraction; TR, tricuspid regurgitation; VC, vena contracta Discussion The PROVE trial showed no significant reduction in RV volumes with carvedilol SR, empagliflozin, or their combination compared with placebo. Although no significant change in TR severity was observed with carvedilol SR, empagliflozin significantly reduced vena contracta width of TR, EROA and regurgitant volume. In the present study, different imaging modalities were used to assess RV volumes and TR severity: CMR for RV volumes and echocardiography for TR. Because CMR is considered the gold standard for quantitative assessment of RV volumes due to the complex shape of the RV and excellent endocardial resolution [19, 21], CMR was used for volumetric analysis of the RV in this trial. CMR is also feasible for quantitative evaluation of TR; however, atrial fibrillation is often associated with functional TR and may pose a challenge for CMR quantification of TR, since most CMR acquisitions are obtained over multiple cardiac cycles [19]. In contrast, echocardiographic assessment of TR has been the standard approach to determine the severity of TR [5, 19], and recent trials involving transcatheter repair and transcatheter valve replacement for TR used echocardiography as the primary imaging modality [22, 23]. Functional TR is common and usually silent in patients with HF [21], and echocardiography is highly useful for evaluating functional TR because of its accessibility, real-time imaging capabilities, and cost-effectiveness, making it the diagnostic modality of choice for evaluation of TR [21, 24]. We set the change in severity of TR as a secondary outcome and included measurement of vena contracta width, EROA, and regurgitant volume in secondary end points, because there was less experience with quantitation of TR severity with PISA compared with mitral regurgitation (MR) [25, 26]. No single Doppler measurement or parameter is precise enough to quantify TR severity; however, TR severity can be determined with high probability, when multiple parameters are concordant [19]. In the present study, the SGLT2 inhibitor, empagliflozin consistently reduced multiple parameters of TR severity significantly. The current guidelines provide no class I recommendation for medical management of functional TR, and diuretic therapy alone has a class IIa recommendation for right-sided HF attributable to severe TR [5]. Functional TR and MR are common in HF with preserved EF (HFpEF), and relate to underlying atrial myopathy of the left atrium and/or right atrium [21, 27]. While use of beta-blockers is not recommended for HFpEF, SGLT2 inhibitors can be beneficial in HFpEF based on remarkable outcomes of recent clinical trials [6, 13, 14], and may be considered for pharmacological management of functional TR and MR associated with HFpEF. In the EFFORT trial (Ertugliflozin for Functional Mitral Regurgitation), we previously demonstrated that compared with placebo, the SGLT2 inhibitor ertugliflozin significantly reduced functional MR associated with HF with mildly reduced EF, although there were no significant between-group differences regarding the changes in LV volume indices and LVEF [26]. The PROVE trial presently showed that the SGLT2 inhibitor empagliflozin showed no significant reduction in RV volumes compared with placebo, but improved functional TR in patients with HFpEF. This study could not determine the mechanism underlying the effect of empagliflozin on TR severity, and further studies are needed to evaluate the way in which empagliflozin works for the reduction of functional TR. Management of severe functional TR should be performed by a multidisciplinary team and optimization of medical therapy is required before consideration of transcatheter therapies for TR [3, 21]. Despite current recommendations of SGLT2 inhibitors for HFpEF, SGLT2 inhibitors are rarely used in patients with functional TR and HFpEF, because their effects on functional TR are uncertain [22]. In this regard, the EVENT trial (Enavogliflozin Outcome Trial in Functional Tricuspid Regurgitation, NCT06027307) will test the hypothesis that the SGLT2 inhibitor enavogliflozin, compared with placebo, will improve clinical and echocardiographic outcome in functional TR associated with HFpEF. This ongoing trial, to be completed in 2027, will document the impact of enavogliflozin versus matching placebo on the occurrence of a composite of cardiovascular events or worsening of TR in 540 patients with functional TR and HFpEF. Limitations Because enrollment of trial patients was affected by the COVID-19 pandemic, the present trial did not reach the calculated sample size to provide 80% power to detect a difference of 15% in the RV end-systolic volume index between the treatment groups, thereby resulting in an underpowered study. The primary analysis did not follow the intention-to-treat principle and instead included only patients who completed the study, owing to the small number of patients in this trial. In this trial, matching placebo for empagliflozin was not provided. However, the assessors were blinded to patient data, treatment allocation, study sequence, and previous measurements in each patient. The daily dose of carvedilol SR used in this trial was lower than the target dose recommended for heart failure with reduced ejection fraction [6]. Our trial enrolled patients with HFpEF, for whom no established target dose of carvedilol exists and the use of beta-blockers for managing RV failure is not well established. Therefore, the lack of efficacy of carvedilol SR may be dose-related rather than reflecting a true class effect of beta-blockers. Because our trial included patients with an LVEF ≥ 50%, the results of our study may not be applicable to patients with LVEF < 50%. Finally, our small trial is a mechanistic study, not an outcome trial, and the statistical analyses were performed according to the prespecified study protocol without formal adjustment for multiple comparisons. Conclusion Among patients with significant functional TR, carvedilol SR, empagliflozin, or their combination showed no significant reduction in RV volumes compared with placebo. However, empagliflozin significantly reduced functional TR. A future outcome trial will determine the efficacy of SGLT2 inhibitors on functional TR secondary to HFpEF. Acknowledgements Not applicable. Abbreviations Sustained-release carvedilol Confidence interval Cardiac magnetic resonance Effective regurgitant orifice area Guideline-directed medical therapy Heart failure Left ventricle Left ventricular ejection fraction Mitral regurgitation New York Heart Association Proximal isovelocity surface area Right ventricle Right ventricular ejection fraction Sodium-glucose cotransporter 2 Tricuspid regurgitation Authors’ contributions Conceptualization: Kang DH. Methodology: Kang DH. Investigation: Kang DH, Kim GY. Formal analysis: Kang DH, Yang DH, Lee JE, Yun SC, Kim GY, Lim SM. Supervision: Kang DH, Song JM, Kim HK, Park SJ, Shim CY, Lee SM, Hwang IC, Yoon YE. Writing—original draft: Kang DH, Kim GY. Writing—review & editing: Kang DH, Song JM, Kim HK, Park SJ, Shim CY, Lee SM, Hwang IC, Yoon YE, Kim GY. Funding This trial was supported in part by a research grant from the investigator-initiated studies program of Chong Kun Dang Pharmaceutical Corporation. The funder provided the trial drugs, placebos, part of funding, but had no role in data collection, data analysis, data interpretation or writing of the manuscript. Data availability The data collected for this study will not be made available to others. Declarations Ethics approval and consent to participate The study protocol was designed by the principal investigator and approved by the institutional review board at each participating center: [Asan Medical Center] (IRB No. 2020–0127), [Samsung Medical Center] (IRB No. 2021–01-016), [Seoul National University Bundang Hospital] (IRB No. B-2008–633-005), [Severance Medical Center] (IRB No. 4–2021-0428), [Seoul National University Hospital] (IRB No. H-2011–185-1177). All the participants provided written informed consent. Consent for publication Not applicable. Competing interests Sung-Ji Park, In-Chang Hwang, Chi Young Shim, and Hyung-Kwan Kim are editorial board members of this journal, but were not involved in the peer review or the decision-making process of this manuscript. The authors declare that they have no other competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The data collected for this study will not be made available to others.