Impact of RAS inhibitors on 2-year mortality and disease progression in chronic mitral regurgitation

Abstract

Background

Chronic mitral regurgitation (MR) is associated with adverse clinical outcomes, and the role of renin-angiotensin system inhibitors (RASI) in patients with MR remains uncertain. This study evaluated the association between RASI use patterns and 2-year all-cause mortality and disease progression in patients with primary MR (PMR) and secondary MR (SMR).

Methods

This registry-based study included 3,297 patients with moderate-to-severe chronic MR from the China Valvular Heart Disease registry, including 955 with PMR and 2,342 with SMR. RASI use was assessed at baseline and during follow-up, and patients were categorized as always users, never users, or inconsistent users. The primary endpoint was 2-year all-cause mortality. The secondary endpoint was change in MR severity. Associations between RASI use and outcomes were assessed using multivariable Cox regression, propensity-score matching, and time-varying Cox analyses.

Results

Consistent RASI use was associated with lower 2-year all-cause mortality in both PMR and SMR. In multivariable Cox regression, never users had higher mortality risk than always users in both cohorts (PMR: HR 5.40, 95% CI 1.58–18.47, P = 0.007; SMR: HR 5.09, 95% CI 2.85–9.09, P < 0.001). Inconsistent users also had higher mortality risk than always users (PMR: HR 7.56, 95% CI 2.25–25.37, P = 0.001; SMR: HR 3.29, 95% CI 1.85–5.84, P < 0.001). Propensity-score-matched analyses showed broadly consistent results. Time-varying Cox analyses demonstrated directionally consistent associations in the overall cohort and SMR during selected follow-up intervals, whereas in PMR the interval-specific estimates were less stable and did not reach statistical significance. Consistent RASI use was also associated with improvement in MR severity, particularly in SMR and in patients without guideline-based indications for valvular intervention.

Conclusions

In patients with moderate-to-severe chronic MR, consistent RASI use was associated with lower 2-year all-cause mortality and improvement in MR severity. These findings suggest that RASI therapy may have a role in optimized medical management for selected patients with chronic MR, particularly those with SMR or impaired/borderline LV function. Further prospective studies are needed to confirm whether RASI directly improves survival or delays MR progression.

Trial registration

This study utilized data from the China Valvular Heart Disease (China-VHD) registry (NCT03484806).

Supplementary Information

The online version contains supplementary material available at 10.1186/s12872-026-06057-4.

Background

Chronic mitral regurgitation (MR) is a prevalent valvular heart disease (VHD) associated with considerable morbidity and mortality. It is the leading type of VHD in China [1, 2] and the second most common in Western countries [3–5]. MR is broadly classified into primary MR (PMR), caused by structural mitral valve abnormalities, and secondary MR (SMR), which arises in the absence of intrinsic valve pathology and may result from ventricular remodeling/dysfunction or atrial pathology. Despite advancements in surgical and transcatheter therapies, the clinical management of MR remains suboptimal. In PMR, mitral valve surgery is the standard for severe, symptomatic cases or those with reduced LV systolic function [6] (LVEF ≤ 60%, LVESD ≥ 40 mm, or LVESD ≥ 20 mm/m²). However, many eligible patients do not receive timely intervention due to delayed referral, late-stage disease at presentation, or limited access to advanced healthcare in resource-constrained settings. In SMR, guideline-directed medical therapy (GDMT) for heart failure (HF) is the cornerstone, with invasive treatment reserved for patients with persistent symptoms and LV dysfunction (LVEF < 50%) [7, 8].

Renin-angiotensin system inhibitors (RASI) are commonly prescribed in patients with MR, particularly when systemic hypertension, left ventricular dysfunction, or concomitant heart failure is present [9–11]. Beyond their established role in neurohormonal modulation, RAS inhibition may plausibly influence MR progression through several mechanisms. By reducing afterload and ventricular wall stress, RASI may decrease regurgitant volume in functional MR. By attenuating activation of the renin–angiotensin–aldosterone system, they may also limit adverse ventricular remodeling, including left ventricular dilatation and annular enlargement, thereby reducing leaflet malcoaptation and the persistence of the regurgitant orifice. In addition, favorable effects on ventricular geometry and loading conditions may contribute to a lower regurgitant fraction even in some patients with PMR. However, whether these mechanisms translate into reduced MR severity or improved survival in patients with chronic MR, either with or without overt heart failure, remains unclear.

Given the lack of randomized evidence and the limited real-world data on longitudinal RASI exposure in chronic MR, a registry-based analysis may provide clinically relevant insight into treatment patterns and outcome associations in routine practice. This study aims to evaluate the association between RASI use and 2-year all-cause mortality and disease progression in patients with moderate-to-severe chronic MR, independent of conventional heart failure indications.

Methods

Study population

This retrospective study utilized data from the China Valvular Heart Disease (China-VHD) registry (NCT03484806), a national, multicenter, prospective cohort study conducted across 46 health centers in mainland China between April and June 2018. The China-VHD registry database has been previously described in detail, with efforts to ensure the accuracy and completeness of the data [12, 13]. For this study, patients with moderate to severe MR, confirmed by echocardiographic evaluation, were included. Patients with other significant valvular diseases (including aortic stenosis, aortic regurgitation, mitral stenosis, tricuspid stenosis, pulmonary valve stenosis, and pulmonary valve regurgitation) were excluded. Concomitant moderate-to-severe tricuspid regurgitation was not an exclusion criterion to better capture the progression of heart failure associated with left-sided valvular disease [14]. Additional exclusion criteria were malignant tumors, previous valvular interventions, deaths within 30 days of enrollment (to exclude patients with acute decompensation or perioperative mortality who were unlikely to have a stable medication regimen and would introduce early-event bias), or incomplete follow-up data.

A total of 4,076 patients with moderate to severe MR were identified. Subsequently, 778 patients who underwent transcatheter or surgical valvular intervention within 30 days of enrollment were excluded. Patients were censored at the time of valvular intervention or surgery during the 2-year follow-up to evaluate the net effect of RASI therapy on long-term all-cause mortality. During the follow-up period, only one patient in the PMR cohort underwent valvular surgery, and no patients in the SMR cohort underwent intervention. These exclusions accounted for 65.9% of PMR and 27.5% of SMR patients who met guideline-based Class I and IIa indications for mitral valve intervention or surgery (Online Supplementary Table S1). Ultimately, a total of 3,297 patients were included in the final analysis, consisting of 955 patients with PMR and 2,342 patients with SMR. Among them, 1,312 patients (39.8%) had concomitant moderate-to-severe TR. The study was approved by the Ethics Committee of Fuwai Hospital and the institutional review board of the National Center for Cardiovascular Diseases of China, with written informed consent obtained from all participants. The flowchart of the study population is shown in Fig. 1.

Fig. 1.

Patient inclusion flow chart. The China-VHD study enrolled 13,917 patients with moderate to severe valvular heart disease confirmed by echocardiographic evaluation. For this analysis, we excluded patients with moderate or greater involvement of other valvular diseases or mixed valvular disease (while secondary tricuspid regurgitation was not excluded), those with malignant tumors, previous valvular interventions, deaths within 30 days, and those without follow-up information, resulting in 4,076 patients with moderate to severe mitral regurgitation (MR). After further excluding patients who underwent valvular intervention or surgery within 30 days of enrollment or during the 2-year follow-up period, a total of 3,297 patients were included in the final analysis, comprising 955 patients with primary MR (PMR) and 2,342 patients with secondary MR (SMR). AR, aortic regurgitation; AS, aortic stenosis; MS, mitral stenosis; PR, pulmonic regurgitation; PS, pulmonic stenosis; TS, tricuspid stenosis

Echocardiographic assessment

Comprehensive two-dimensional transthoracic echocardiography was performed for all patients using commercially available ultrasound systems. Left ventricular ejection fraction (LVEF) was calculated using the biplane-modified Simpson method, in accordance with the recommendations of current guidelines [15]. Senior doctors determined the etiology of VHDs in combination with echocardiographic features and medical history. Pulmonary hypertension was defined as an estimated resting systolic pulmonary arterial pressure (sPAP) > 50 mmHg based on Doppler echocardiography, consistent with current guideline thresholds for consideration of intervention in valvular heart disease. The severity of MR and tricuspid regurgitation was assessed based on echocardiographic valve hemodynamic parameters, as previously described in published literature [12, 13].

Patterns of RASI use

RASI medications included angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs). Angiotensin receptor-neprilysin inhibitors (ARNIs) were not included because of their limited availability in China during the study period. Adherence to RASI therapy was assessed over a uniform 2-year follow-up period after enrollment. RASI use status was recorded at 5 assessment time points: baseline, 6 months, 12 months, 18 months, and 24 months. According to RASI use patterns across these 5 time points, patients were categorized into 3 groups: always users, who reported RASI use at baseline and at all follow-up visits; never users, who reported no RASI use at baseline or any follow-up visit; and inconsistent users, who showed non-persistent or irregular exposure patterns that did not satisfy either of the above definitions. Detailed trajectory patterns among inconsistent users, including discontinuation after baseline use, late initiation with subsequent persistence, and exposure-intensity categories, are presented in Online Supplementary Table S2.

Indications for valvular intervention

Patients were classified as having or not having an indication for mitral valve intervention according to Class I or IIa recommendations from contemporary ACC/AHA and ESC/EACTS valvular heart disease guidelines [7, 8]. For PMR, intervention indications included symptomatic severe PMR and asymptomatic severe PMR with evidence of left ventricular systolic dysfunction or enlargement, such as LVEF ≤ 60% or left ventricular end-systolic diameter (LVESD) ≥ 40 mm. For SMR, intervention indications included carefully selected patients with severe symptoms and reduced LVEF (< 50%) despite guideline-directed medical therapy, as well as patients with moderate PMR or SMR undergoing cardiac surgery for other indications such as coronary artery bypass grafting.

Endpoints

The primary endpoint was 2-year all-cause mortality. The secondary endpoint was change in MR severity during follow-up. Follow-up was conducted every 6 months for 2 years through outpatient visits, medical records, or telephone interviews. Reported deaths and valve interventions were reviewed and verified by investigators at each participating center.

Statistical analysis

Continuous variables were summarized according to their distributional characteristics, using mean ± standard deviation (SD) for approximately symmetric variables and median with interquartile range (IQR) when appropriate. Categorical variables were expressed as number (percentage). Group comparisons were performed using Student’s t-test or Wilcoxon rank-sum test for continuous variables, and chi-square tests for categorical variables.

Missing data were handled according to the analytic model. Baseline covariates included in the primary analyses were complete or had negligible missingness (< 1%); therefore, multiple imputation was not applied in the primary Cox analyses. Multiple imputation was applied only in the pre-specified time-varying Cox sensitivity analysis to account for missingness in longitudinal RASI exposure. Twenty imputed datasets were generated using the mice package in R, and estimates were combined according to Rubin’s rules.

Survival analyses were performed using Kaplan–Meier curves with log-rank testing. Associations between RASI use patterns and 2-year all-cause mortality were evaluated using multivariable Cox proportional-hazards regression. Prespecified baseline covariates included age, sex, body mass index, hypertension, hypercholesterolemia, diabetes mellitus, coronary artery disease, prior myocardial infarction, prior percutaneous coronary intervention, prior coronary artery bypass grafting, atrial fibrillation, stroke, chronic kidney disease, heart failure, systolic blood pressure, diastolic blood pressure, heart rate, New York Heart Association (NYHA) class III/IV, left atrial diameter, LVESD, left ventricular end-diastolic diameter (LVEDD), LVEF, pulmonary hypertension, MR severity, significant TR, and baseline background medications including beta-blockers, diuretics, anticoagulants, antiplatelet therapy, and digoxin. Hazard ratios (HRs) and 95% confidence intervals (CIs) were reported.

Change in MR severity between baseline and follow-up was quantified according to change in transthoracic echocardiographic MR grade. Improvement was defined as a decrease in MR severity grade at follow-up relative to baseline. Because MR severity was analyzed as a repeated within-patient multi-category variable, paired comparisons were assessed using the Stuart-Maxwell test.

Sensitivity analyses were performed to assess robustness. First, propensity score matching (PSM) was conducted as a sensitivity analysis using 1:1 nearest-neighbor matching without replacement. Covariate balance after matching was assessed using standardized mean differences (SMDs), and all-cause mortality was further analyzed in the matched cohorts. Second, time-varying Cox analyses were performed in which RASI exposure was modeled as a time-updated covariate across prespecified intervals (0–6, 6–12, 12–18, and 18–24 months). Subgroup analyses were performed in predefined strata, and interaction was assessed using P for interaction rather than routine inclusion of interaction terms in the primary models. All tests were 2-sided, and P < 0.05 was considered statistically significant. Analyses were performed using SPSS version 26.0 and R version 4.2.2.

A detailed description of the study methods is available in the Online Supplementary Extended methods.

Results

Baseline characteristics and independent predictors of mortality

A total of 3,297 patients with moderate-to-severe chronic MR were analyzed, including 955 patients (29.0%) with PMR and 2,342 patients (71.0%) with SMR. 1,312 patients (39.8%) had concomitant secondary moderate-to-severe tricuspid regurgitation. The specific patient selection flowchart was depicted in Fig. 1. The average age of patients in the PMR cohort was 67.9 ± 12.3 years, while the SMR cohort had a younger average age of 64.4 ± 12.4 years (P < 0.001). Baseline characteristics differed significantly between the PMR and SMR cohorts. Patients with PMR were more likely to have hypertension and atrial fibrillation, whereas patients with SMR exhibited higher comorbidity rates of coronary artery disease, hypercholesterolemia, diabetes mellitus, and heart failure. Echocardiographic parameters also revealed notable differences, with SMR patients exhibiting larger left ventricular end-systolic diameter (LVESD) and left ventricular end-diastolic diameter (LVEDD) and lower LVEF. Despite these differences, left atrial diameter (LAD), pulmonary hypertension, and MR severity were comparable between the groups. Baseline demographic, clinical, echocardiographic, and background medication characteristics are summarized in Table 1.

Table 1.

Baseline demographic and clinical characteristics in the overall cohort

Variables	Overall (n = 3297)	PMR (n = 955)	SMR (n = 2342)	P Value
Demographics
Male sex	1950 (59.1)	487 (51.0)	1436 (62.5)	< 0.001
Age, years	65.4 ± 12.5	67.9 ± 12.3	64.4 ± 12.4	< 0.001
BMI, kg/m2	23.9 ± 3.6	23.5 ± 3.4	24.1 ± 3.6	< 0.001
Medical history
Smoker	1211 (36.7)	261 (27.3)	950 (40.6)	< 0.001
Hypertension	1675 (50.8)	513 (53.7)	1162 (49.6)	0.036
Hypercholesterolemia	534 (16.2)	122 (12.8)	412 (17.6)	< 0.001
Diabetes	764 (23.2)	192 (20.1)	572 (24.4)	0.009
Coronary artery disease	1348 (40.9)	340 (35.6)	1008 (43.0)	< 0.001
Acute myocardial infarction	559 (17.0)	106 (11.1)	453 (19.3)	< 0.001
Percutaneous coronary intervention	601 (18.2)	141 (14.8)	460 (19.6)	0.001
Coronary artery bypass graft	69 (2.1)	17 (1.8)	52 (2.2)	0.505
Atrial fibrillation	988 (30.0)	328 (34.3)	660 (28.2)	< 0.001
Stroke	329 (10.0)	110 (11.5)	219 (9.4)	0.069
Chronic kidney disease	285 (8.6)	77 (8.1)	208 (8.9)	0.490
Heart failure	2151 (65.2)	547 (57.3)	1604 (68.5)	< 0.001
Background medications
Beta-blocker	1924 (58.4)	477 (49.9)	1447 (61.8)	< 0.001
Diuretic	1503 (45.6)	383 (40.1)	1120 (47.8)	< 0.001
Anticoagulant	652 (19.8)	229 (24.0)	423 (18.1)	< 0.001
Antiplatelet therapy	1524 (46.2)	405 (42.4)	1119 (47.8)	0.006
Digoxin	630 (19.1)	188 (19.7)	442 (18.9)	0.624
Clinical condition
SBP, mmHg	125.8 ± 21.8	128.1 ± 21.3	124.9 ± 21.9	< 0.001
DBP, mmHg	76.6 ± 13.6	76.2 ± 13.2	76.9 ± 13.8	0.174
Heart rate, bpm	80.2 ± 19.2	79.6 ± 19.4	80.4 ± 19.1	0.283
NYHA classification				0.004
I	39 (1.8)	11 (2.0)	28 (1.7)
II	610 (28.4)	187 (34.2)	423 (26.4)
III	985 (45.8)	236 (43.1)	749 (46.7)
IV	517 (24.0)	113 (20.7)	404 (25.2)
Echocardiography
LAD, mm	46.6 ± 8.2	46.4 ± 8.7	46.8 ± 8.1	0.199
LVESD, mm	42.9 ± 10.4	39.4 ± 9.0	44.4 ± 10.6	< 0.001
LVEDD, mm	57.8 ± 10.4	54.3 ± 9.1	59.3 ± 10.6	< 0.001
LVEF, %	45.8 ± 14.9	52.2 ± 13.4	43.3 ± 14.6	< 0.001
Pulmonary hypertension	1492 (45.3)	431 (45.1)	1061 (45.3)	0.959
Mitral regurgitation severity				0.078
Moderate	2335 (70.8)	655 (68.6)	1680 (71.7)
Severe	962 (29.2)	300 (31.4)	662 (28.3)
Tricuspid regurgitation severity				0.368
None or mild	1985 (60.2)	563 (59.0)	1422 (60.7)
Moderate to severe	1312 (39.8)	392 (41.0)	920 (39.3)

Data are presented as mean ± SD, median (interquartile range), or number (%)

BMI Body mass index, DBP Diastolic blood pressure, LAD Left atrium diameter, LVEDD Left ventricular end-diastolic diameter, LVEF Left ventricular ejection fraction, LVESD Left ventricular end-systolic diameter, PMR Primary mitral regurgitation, RASI Renin-angiotensin system inhibitors, SBP Systolic blood pressure, SMR Secondary mitral regurgitation

During the 2-year follow-up, a total of 103 deaths (10.7%) were recorded in the PMR cohort and 232 deaths (9.9%) in the SMR cohort, with no statistically significant difference between the two groups (P = 0.921). Kaplan–Meier survival curves illustrate the survival distributions across the entire study population (Online Supplementary Figure S1). Candidate baseline risk factors included in the mortality analyses are listed in Online Supplementary Table S3. Multivariable Cox regression analysis identified age and New York Heart Association class III/IV as independent predictors of all-cause mortality in both cohorts. Baseline background medications, including beta-blockers, diuretics, anticoagulants, and antiplatelet therapy, were also independently associated with mortality in both PMR and SMR. In PMR, diabetes, chronic kidney disease, and systolic blood pressure remained significant in the multivariable model, whereas in SMR, hypercholesterolemia, heart rate, and eGFR remained significant. The detailed univariable and multivariable Cox results for PMR and SMR are presented in Online Supplementary Tables S4 and S5, respectively. The multivariable models showed good discriminative performance in both cohorts, with C-index values of 0.824 for PMR and 0.823 for SMR (both P < 0.001).

Effect of RASI use patterns on survival outcomes

The combined proportion of always and inconsistent users constituted 60.1% (n = 574) of the PMR cohort and 68.2% (n = 1,597) of the SMR cohort (P < 0.001). Detailed trajectory patterns among inconsistent RASI users, including baseline use followed by discontinuation, late initiation, and different exposure-intensity categories, are shown in Online Supplementary Table S2. Baseline characteristics stratified by RASI use patterns are presented in Online Supplementary Tables S6 and S7. In both PMR and SMR cohorts, patients with consistent RASI use (always users) exhibited a higher burden of cardiovascular comorbidities compared with never users, particularly hypertension and heart failure. For example, in the PMR cohort, the prevalence of hypertension increased from 40.2% in never users to 69.1% in always users, and heart failure from 51.2% to 65.9% (both P < 0.01). Similar patterns were observed in the SMR cohort. Echocardiographic parameters also differed across groups, with always users demonstrating more advanced left ventricular remodeling, including larger LVESD and LVEDD and lower LVEF compared with never users (all P < 0.05), suggesting a more severe baseline disease profile. Inconsistent users generally showed intermediate characteristics between the two groups.

At the 2-year follow-up, consistent RASI use was associated with significantly higher survival rates compared with never users in both PMR and SMR cohorts (PMR: 95.6% vs. 87.6%; SMR: 97.4% vs. 81.5%; both P < 0.001), as illustrated by the Kaplan–Meier survival curves (Fig. 2). Notably, inconsistent users exhibited intermediate survival rates, which were significantly better than those of never users in the SMR cohort (86.8% vs. 81.5%; P < 0.05) but not in the PMR cohort (83.8% vs. 87.6%; P = 0.12).

Fig. 2.

Kaplan–Meier curve of overall survival by RASI use patterns. The rate of overall survival at 2-year follow-up among different RASI treatment patterns in PMR (A) and SMR (B) was presented as mean ± SD. The hazard ratio (HR) and 95% confidence interval (CI) were calculated with never users as reference. PMR, primary mitral regurgitation; RASI, renin-angiotensin system inhibitors; SMR, secondary mitral regurgitation

Multivariable Cox proportional-hazards regression showed that, compared with always users, never users had a significantly higher risk of all-cause mortality in both cohorts (PMR: HR 5.40, 95% CI 1.58–18.47, P = 0.007; SMR: HR 5.09, 95% CI 2.85–9.09, P < 0.001). Inconsistent users also had a significantly higher mortality risk than always users (PMR: HR 7.56, 95% CI 2.25–25.37, P = 0.001; SMR: HR 3.29, 95% CI 1.85–5.84, P < 0.001). In PMR, the mortality risk of never users did not differ significantly from that of inconsistent users (HR 0.71, 95% CI 0.46–1.11, P = 0.133), whereas in SMR, never users had a significantly higher risk than inconsistent users (HR 1.55, 95% CI 1.16–2.07, P = 0.003) (Table 2).

Table 2.

Long-term risk of death according to RASI use pattern in PMR and SMR cohorts

	Multivariable Cox Model			PS Model
	Sample size	HR (95% CI)	P value	Sample size	HR (95% CI)	P value
PMR
Never vs. Always	381 vs 223	5.40 (1.58–18.47)	0.007	153 vs 153	12.59 (1.64–96.82)	0.015
Inconsistent vs. Always	351 vs 223	7.56 (2.25–25.37)	0.001	183 vs 183	10.69 (2.53–45.23)	0.001
Never vs. Inconsistent	381 vs 351	0.71 (0.46–1.11)	0.133	276 vs 276	0.75 (0.48–1.16)	0.199
SMR
Never vs. Always	745 vs 763	5.09 (2.85–9.09)	< 0.001	475 vs 475	5.38 (2.64–10.96)	< 0.001
Inconsistent vs. Always	834 vs 763	3.29 (1.85–5.84)	< 0.001	601 vs 601	5.28 (2.62–10.62)	< 0.001
Never vs. Inconsistent	745 vs 834	1.55 (1.16–2.07)	0.003	555 vs 555	1.53 (1.11–2.10)	0.009

In multivariable Cox model, proportional-hazards regression was used with adjustment for all patient-level variables in Table S3. In PS model, a logistic regression was performed to estimate propensity score, adjusting for all patient-level variables in Table S3. We established 1:1 PS matching cohort based on PS score. Matching specifications were shown in Table S8-S9, Online Supplement. HRs were reported according to the comparison shown in each row

CI Confidence interval, HR Hazard ratio, PMR Primary mitral regurgitation, PS Propensity score, SMR Secondary mitral regurgitation

Sensitivity analyses

In the propensity-score-matched cohorts, the overall direction of association remained broadly consistent with the primary multivariable Cox analyses. Matching balance and baseline characteristics after propensity-score matching are shown in Online Supplementary Tables S8 and S9 for PMR and SMR, respectively. In PMR, compared with always users, never users (HR 12.59, 95% CI 1.64–96.82; P = 0.015) and inconsistent users (HR 10.69, 95% CI 2.53–45.23; P = 0.001) both had higher mortality risk, whereas never users did not differ significantly from inconsistent users (HR 0.75, 95% CI 0.48–1.16; P = 0.199). In SMR, never users (HR 5.38, 95% CI 2.64–10.96; P < 0.001) and inconsistent users (HR 5.28, 95% CI 2.62–10.62; P < 0.001) also had significantly higher mortality risk than always users, and never users had a higher mortality risk than inconsistent users (HR 1.53, 95% CI 1.11–2.10; P = 0.009). These matched-cohort outcome results are included in Table 2.

Time-varying Cox analyses showed that the association between RASI use and all-cause mortality varied across follow-up intervals. In the overall cohort, RASI use was associated with lower all-cause mortality during 0–6 months (HR 0.68, 95% CI 0.48–0.95; P = 0.026) and 18–24 months (HR 0.47, 95% CI 0.25–0.90; P = 0.022), whereas no significant association was observed during 6–12 months or 12–18 months. In PMR, no interval showed a statistically significant association. In SMR, RASI use was associated with lower mortality during 0–6 months (HR 0.51, 95% CI 0.34–0.78; P = 0.002) and 18–24 months (HR 0.38, 95% CI 0.18–0.83; P = 0.015), while the intermediate intervals were not statistically significant. The interval-specific results are shown in Fig. 3, and the full time-varying Cox model results are provided in Online Supplementary Table S10.

Fig. 3.

Time-varying association between RASI use and all-cause mortality across follow-up intervals. Time-varying Cox proportional-hazards models were used to estimate the association between updated RASI use and all-cause mortality across prespecified follow-up intervals (0–6, 6–12, 12–18, and 18–24 months) in the overall cohort (A), the PMR cohort (B), and the SMR cohort (C). Hazard ratios (HRs) and 95% confidence intervals (CIs) are shown for RASI use within each interval. Models were adjusted for baseline demographic, clinical, echocardiographic, and background medication variables. CI, confidence interval; HR, hazard ratio; PMR, primary mitral regurgitation; RASI, renin-angiotensin system inhibitor; SMR, secondary mitral regurgitation

Subgroup analysis and impact of RASI use on MR severity

Subgroup analyses suggested that the association between consistent RASI use and lower mortality was generally observed across major clinical subgroups in both PMR and SMR, including sex, hypertension, heart failure, MR severity, significant TR, and guideline-based indications for intervention, although not all subgroup estimates reached statistical significance. Patients with LVEF < 50% or 50–60% appeared to derive greater survival benefit from consistent RASI use than those with LVEF > 60%; however, interaction tests were not statistically significant (PMR: P for interaction = 0.113; SMR: P for interaction = 0.307). Additional etiologic subgroup analyses in SMR, including ischemic and functional SMR, are also shown. These subgroup findings are presented in Fig. 4.

Fig. 4.

Hazard ratios associated with RASI treatment in subgroups. Subgroup analyses were performed using Cox proportional-hazards regression models in the PMR (A) and SMR (B) cohorts. The HRs and 95% CIs are reported for inconsistent users or never users, with always users as the reference group. Additional etiologic subgroup analyses within the SMR cohort are shown in panel C, including ischemic SMR and functional SMR. CI, confidence interval; HF, heart failure; HR, hazard ratio; MR, mitral regurgitation; PMR, primary mitral regurgitation; RASI, renin-angiotensin system inhibitor; SMR, secondary mitral regurgitation; TR, tricuspid regurgitation

Changes in MR severity from baseline to 2-year follow-up were assessed using repeated echocardiographic measurements. Consistent RASI use was associated with a greater proportion of MR severity improvement, more evidently in the SMR cohort. Changes in MR severity according to RASI use, intervention indication status, and MR etiology are shown in Fig. 5.

Fig. 5.

Mitral regurgitation severity changes at 2-year follow-up compared to baseline. Changes of MR severity were compared between never and always RASI users in the overall cohort (left panel), and in patients with (middle panel) and without (right panel) transcatheter or surgical indications for mitral valve intervention in PMR (A) and SMR (B) cohorts. PMR, primary mitral regurgitation; RASI, renin-angiotensin system inhibitors; SMR, secondary mitral regurgitation

Discussion

This large registry-based study evaluated the association between RASI use patterns and clinical outcomes in patients with moderate-to-severe chronic MR, separately in PMR and SMR cohorts. Consistent RASI use was associated with lower 2-year all-cause mortality in both cohorts, with a more evident association among patients with LVEF < 50% or 50–60%, although interaction tests were not statistically significant. Time-varying Cox analyses showed directionally consistent associations in the overall cohort and in SMR during selected follow-up intervals, whereas the association was less stable in PMR. Furthermore, consistent RASI use was also associated with a greater proportion of MR severity improvement, although these echocardiographic findings remain exploratory. These findings suggest that RASI therapy may have clinical relevance in the medical management of chronic MR, while the observed associations should be interpreted cautiously given the registry-based observational design.

Bridging gaps in current guidelines for SMR

SMR is a heterogeneous condition arising from a variety of underlying mechanisms. Ventricular SMR is commonly related to ischemic heart disease or non-ischemic cardiomyopathy, in which LV remodeling, papillary muscle displacement, and annular dilatation contribute to leaflet malcoaptation. In addition, atrial functional MR is increasingly recognized as a distinct SMR phenotype, particularly in patients with atrial fibrillation or marked left atrial enlargement, in whom mitral annular dilatation may occur despite relatively preserved LV systolic function. These different mechanisms highlight the importance of optimizing medical therapy according to the underlying cardiac substrate.

Optimization of GDMT is the cornerstone of SMR management, and invasive intervention is generally considered only after appropriate medical therapy has been optimized [7, 8]. More recent trials, such as MITRA-FR and COAPT, have highlighted the critical role of SMR in worsening HF outcomes and emphasized the importance of optimizing GDMT before considering invasive interventions [16, 17]. RASI therapy is a key component of GDMT in heart failure, particularly in patients with reduced LVEF. Early landmark heart failure trials established the survival benefit of ACEIs and ARBs in patients with HFrEF [18–22]. More recent studies have further supported the biological plausibility that neurohormonal inhibition may improve MR-related outcomes through reverse remodeling. The PRIME trial demonstrated that sacubitril/valsartan reduced the effective regurgitant orifice area and regurgitant volume of functional MR compared with valsartan alone [23]. In addition, an analysis from the PROVE-HF study showed that initiation of sacubitril/valsartan was associated with improvement in MR severity in patients with HFrEF [24]. This underscores the importance of RASI therapy as a cornerstone of medical management in SMR, particularly for patients unsuitable for surgical or transcatheter interventions. These findings provide biological plausibility for the associations observed in our study.

Nevertheless, existing evidence has not fully addressed RASI therapy specifically in patients with chronic MR. Many heart failure trials likely included patients with SMR, but outcomes were not generally stratified according to MR status. Valvular heart disease guidelines emphasize GDMT optimization in SMR but provide limited MR-specific recommendations regarding RASI use outside conventional heart failure indications. In current heart failure guidelines, HF-GDMT is primarily guided by LVEF thresholds, with a Class I recommendation for RASI therapy in patients with LVEF ≤ 40% and a Class IIb recommendation for those with LVEF 41–49% [9–11].

This consideration is clinically important because LVEF may overestimate intrinsic LV systolic function in patients with MR, owing to systolic regurgitant flow into the low-impedance left atrium [23]. As a result, some patients with clinically relevant LV dysfunction may not be fully captured by standard LVEF thresholds.

In our study, consistent RASI use was associated with lower 2-year mortality in patients with SMR, and this association was generally observed across several clinically relevant subgroups, including sex, hypertension, MR severity, significant TR, and intervention indication status. The association appeared more pronounced among patients with LVEF < 60%, whereas no clear benefit was observed in those with LVEF > 60%. Consistent RASI use was also associated with improvement in MR severity in the SMR cohort. These findings are consistent with the concept that RASI therapy may be particularly relevant in SMR phenotypes related to ventricular remodeling and heart failure. However, given the heterogeneity of SMR, the potential benefit of RASI should not be interpreted as uniform across all SMR mechanisms, especially in atrial functional MR, where the dominant mechanism may be atrial and annular remodeling rather than LV systolic dysfunction.

Alternative therapeutic options for PMR

PMR differs fundamentally from SMR because the primary abnormality lies in the mitral valve apparatus rather than ventricular or atrial remodeling. Therefore, mitral valve surgery remains the standard treatment for severe PMR, especially in symptomatic patients or in those with evidence of LV systolic dysfunction or enlargement, including LVEF ≤ 60%, LVESD ≥ 40 mm, or LVESD ≥ 20 mm/m². Medical therapy should not be considered a replacement for timely surgical or transcatheter intervention when guideline-based indications are present.

However, under treatment of eligible patients remains an important real-world problem. The CHINA-VHD registry has shown that the intervention rate for severe symptomatic MR (31.5%) is markedly lower compared to other left-sided valvular heart diseases, such as aortic stenosis (62.7%) and aortic regurgitation (57.7%) [1]. Among PMR patients with guideline-based intervention indications in the source registry, 65.9% underwent intervention and were excluded, whereas approximately one-third remained medically managed and were included in the present analysis (Online Supplementary Tables S1). This treatment gap may reflect patient preference for conservative management, advanced age or comorbidities increasing procedural risk, delayed referral, regional differences in access to specialized valve centers, or socioeconomic barriers. Geographic differences in PMR management have also been reported, suggesting that patterns of referral and intervention may differ between Asian and Western practice settings [25]. These unmet treatment needs highlight the practical relevance of exploring effective medical therapies such as RASI as a bridge or alternative to intervention.

Within this context, RASI therapy may have potential value in selected PMR patients, particularly those with hypertension, early LV remodeling, or borderline LV function. In our study, consistent RASI use was associated with lower mortality in the PMR cohort. However, this association was less stable in time-varying analyses than in SMR, which is biologically plausible because PMR is primarily a structural valvular disease rather than a remodeling-driven disease. The apparently higher relative risk observed in inconsistent users compared with never users should not be interpreted as a dose–response relationship. In the PMR cohort, the inconsistent-use group does not represent a moderate-exposure category but rather reflects non-sustained use. As shown in Supplementary Table S2, this group is heterogeneous and includes a substantial proportion of patients who were treated with RASI at baseline but subsequently discontinued therapy. Clinically, such patterns may indicate drug intolerance, hemodynamic deterioration, or worsening overall clinical status; therefore, this group cannot be simply considered an intermediate category between always users and never users within a dose–response framework. Furthermore, PMR is primarily driven by structural valvular pathology, and any potential benefit of RASI is more likely to depend on continuous and stable exposure rather than intermittent use. Accordingly, the observed findings in the PMR cohort are more likely to reflect discontinuation-related bias, group heterogeneity, and residual confounding, rather than a true harmful effect of partial RASI exposure. The direct comparison between never users and inconsistent users in PMR did not reach statistical significance and these findings should be interpreted with caution. Therefore, RASI therapy in PMR should be viewed as a possible adjunctive strategy rather than disease-definitive therapy. Its potential role may be greatest in patients who are not yet indicated for intervention, those awaiting intervention, or those in whom surgery is deferred because of clinical or healthcare-system factors.

The observed association between RASI use and MR severity improvement in PMR should also be interpreted in the context of PMR pathophysiology. RASI may reduce afterload and wall stress, which could transiently reduce regurgitant burden, but it is unlikely to reverse primary leaflet or chordal pathology. Therefore, any potential benefit in PMR is more likely related to hemodynamic optimization, prevention of adverse LV remodeling, or treatment of coexisting cardiovascular conditions rather than direct correction of the mitral valve lesion.

Clinical interpretation

Taken together, these findings support the potential role of RASI therapy as part of optimized medical management in chronic MR, especially in SMR and in patients with evidence of LV dysfunction or remodeling. The findings also suggest that LVEF thresholds alone may not fully capture the population of MR patients who could benefit from neurohormonal inhibition, given the unique loading conditions in MR. At the same time, the results should be interpreted as hypothesis-generating. The large effect sizes in Table 2, particularly in PMR, together with the wide confidence intervals, residual confounding, and time-related bias, require cautious interpretation. RASI therapy may be most clinically relevant as part of comprehensive GDMT in SMR, while in PMR it should be regarded as an adjunctive strategy for selected patients rather than an alternative to guideline-directed valve intervention.

Limitations and future directions

This study has several limitations. First, because this was a registry-based observational study, causality cannot be established and several sources of bias may have influenced the observed associations. RASI therapy was not randomly assigned, and clinicians may have withheld or discontinued RASI because of hypotension, renal dysfunction, hyperkalemia, medication intolerance, frailty, or advanced clinical status. These factors are themselves associated with poor prognosis and may have contributed to confounding by indication. Conversely, patients who consistently used RASI may have had better medication adherence, closer follow-up, or greater access to healthcare, resulting in healthy-user bias. In addition, classification of RASI exposure according to use patterns over the 2-year follow-up may introduce time-related bias, including immortal time bias, because patients categorized as always users had to remain alive and under follow-up long enough to maintain continuous RASI use. To address these concerns, we performed multivariable Cox regression, propensity-score matching, and time-varying Cox analyses; however, residual confounding and time-related bias cannot be fully eliminated. Therefore, the fixed exposure-pattern results should be interpreted cautiously.

Second, the study did not capture detailed information on RASI subtype, dosage, dose titration, or reasons for discontinuation. Therefore, we could not compare ACEIs with ARBs, evaluate dose–response relationships, or determine whether discontinuation was due to adverse events or clinical deterioration. In addition, ARNIs were not widely available in China during the study period and were therefore not included in the exposure definition. Contemporary HF-GDMT also includes MRAs and SGLT2 inhibitors in eligible patients, but these therapies were not consistently available in the final analytic dataset and therefore could not be adequately adjusted for, leaving potential residual confounding by overall GDMT intensity. Although the PRIME study showed that sacubitril/valsartan can reduce functional MR severity, whether this translates into survival benefit in broader MR populations remains uncertain [23].

Third, exclusion or censoring of patients who underwent mitral valve intervention may have introduced selection bias. Patients who deteriorated clinically and subsequently underwent surgery or transcatheter intervention may have been removed from the risk set, potentially affecting the observed associations between RASI use and outcomes. This issue may be particularly relevant in PMR, where patients with guideline-based indications are more likely to undergo valve intervention.

Fourth, changes in MR severity were assessed using paired echocardiographic measurements and should be considered exploratory. Although paired testing was used to account for repeated measurements within the same individuals, multiplicity correction was not performed. In addition, follow-up echocardiographic availability and inter-center variability may have influenced the assessment of MR severity change. Therefore, these findings should be interpreted as supportive evidence of biological plausibility rather than definitive proof that RASI directly reduces MR severity, and subgroup patterns in MR severity change should not be overstated.

Finally, the 2-year follow-up period may be relatively short for evaluating the natural history and long-term progression of chronic MR, especially PMR. Longer-term prospective studies are needed to confirm whether RASI therapy is associated with durable survival benefit or delayed disease progression in different MR phenotypes.

Conclusions

In this nationwide registry-based study, consistent RASI use was associated with lower 2-year all-cause mortality in patients with moderate-to-severe chronic MR. This association was more consistently supported in SMR, whereas in PMR the time-varying results were less stable and should be viewed as hypothesis-generating. RASI use was also associated with improvement in MR severity, particularly among patients with SMR and those without guideline-based indications for valvular intervention; however, these exploratory findings should be interpreted with caution. These findings suggest that RASI therapy may have a role as part of optimized medical management in selected patients with chronic MR. However, given the observational design and potential residual confounding, further prospective studies are needed to confirm whether RASI directly improves survival or delays MR progression.

Abbreviations

ACC/AHA

American College of Cardiology/American Heart Association

ACEI

Angiotensin-converting enzyme inhibitor

ARB

Angiotensin receptor blocker

ARNI

Angiotensin receptor-neprilysin inhibitor

CI

Confidence interval

CMR

Cardiac magnetic resonance

ESC/EACTS

European Society of Cardiology/European Association for Cardio-Thoracic Surgery

GDMT

Guideline-directed medical therapy

HF

Heart failure

HFrEF

Heart failure with reduced ejection fraction

HR

Hazard ratio

IQR

Interquartile range

LAD

Left atrial diameter

LV

Left ventricular

LVEDD

Left ventricular end-diastolic diameter

LVEF

Left ventricular ejection fraction

LVESD

Left ventricular end-systolic diameter

MR

Mitral regurgitation

NYHA

New York Heart Association

PMR

Primary mitral regurgitation

PSM

Propensity score matching

RAS

Renin-angiotensin system

RASI

Renin-angiotensin system inhibitors

SD

Standard deviation

SMR

Secondary mitral regurgitation

sPAP

Systolic pulmonary arterial pressure

TR

Tricuspid regurgitation

VHD

Valvular heart disease

Authors’ contributions

YW, HX, and EZ conceived the study. EZ and CW developed methodology of this study. EZ and CW drafted the initial manuscript. EZ, CW, XH, JL, ZZ, YY, HX, and YW contributed to the data acquisition, analysis and interpretation of data, and revised the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by grants from the National Twelfth Five-year Science and Technology Support Projects by the Ministry of Science and Technology of China [grant number 2015BAI12B02] to Y. Wu and H. Xu, the National Natural Science Foundation of China [grant number 82170378] and the National High Level Hospital Clinical Research Funding [grant number 2022-GSP-QZ-6, 2024-GSP-TJ-7] to E. Zhang, the Beijing Natural Science Foundation [grant number 7244395] and the Chinese Academy of Medical Sciences Innovation Fund for Clinical and Translational Medicine [grant number 2021-I2M-C&T-B-029] to Y. Ye, and the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences [grant number 2017-12 M-3–002] to J. Lv.

Data availability

The data used during the current study are available from the corresponding authors on reasonable request.

Declarations

Ethics approval and consent to participate

Written informed consent was obtained from all participants. The study was approved by the Ethics Committee of Fuwai Hospital and the National Center for Cardiovascular Diseases of China, and was conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.