Effect of β‐blocker on patients with moderate functional mitral regurgitation undergoing surgical aortic valve replacement

Abstract

Aims

The optimal treatment for severe aortic valve disease complicated with moderate function mitral regurgitation (FMR) remains controversial. Although isolated surgical aortic valve replacement (SAVR) is reasonable, previous studies also show that moderate FMR might deteriorate after surgical treatment and result in poorer prognosis. Because the left ventricular remodelling plays a critical role in the development of FMR, these patients might potentially benefit from the administration of β‐blocker (BB). Unfortunately, relevant clinical evidence is lacking. This study aimed to investigate the impact of post‐operative administration of BB on the outcomes of moderate FMR patients undergoing isolated SAVR.

Methods

In this single‐centre cohort study, patients who underwent isolated SAVR and complicated with pre‐operative moderate FMR during 2010 and 2019 at our centre were retrospectively recruited. Patients were divided into two groups according to postoperative administration of BB (BB group vs. control group). The cumulative survival rates were calculated using the Kaplan–Meier method and tested by the log‐rank test, followed by inverse probability treatment weighting (IPTW) analysis to further control the between‐group imbalances. The primary outcome was the major adverse cardiovascular and cerebrovascular events (MACCE), a composite endpoint of all‐cause death, repeat heart valve surgery, non‐fatal myocardial infarction, stroke, and hospitalization for heart failure.

Results

A total of 165 patients were enrolled, 57 (34.6%) of whom were female, and the mean age was 59.2 ± 12.2 years. Eighty (48.5%) patients received post‐operative BB therapy. The median follow‐up time was 18.4 months. The administration of BB was not associated with lower risk of MACCE [hazard ratio (HR): 0.68, 95% confidence interval (CI): 0.29–1.62, P = 0.388] or all‐cause death (HR: 1.03, 95% CI: 0.30–0.56, P = 0.967). In the IPTW dataset, the total number of patients were 326.89, and the outcomes regarding the risk of MACCE (HR: 0.79, 95% CI: 0.31–1.97, P = 0.607) and all‐cause death (HR: 1.33, 95% CI:0.35–5.05, P = 0.674) were in line with the unmatched analysis. The follow‐up echocardiographic results were available for 72.2% of the overall cohort, and the use of BB was observed to be associated with higher improvement rate of follow‐up FMR according to the IPTW analysis (92.2% vs. 98.3%, P = 0.033).

Conclusions

The administration of BB after SAVR was not associated with lower risk of MACCE for patients of severe aortic valve disease complicated with moderate FMR, but was potentially beneficial for improving FMR.

Introduction

Functional mitral regurgitation (FMR) is a common complication among patients with severe aortic valve disease scheduled for surgical aortic valve replacement (SAVR). ¹ Aortic valve diseases contribute to the hypertrophy and enlargement of the left ventricle, which results in the dilation of the mitral annulus, leading to the occurrence of FMR. In other words, left ventricular remodelling plays an important pathogenic role in the development of FMR.

Although mild FMR could be resolved along with the SAVR, severe FMR requires concomitant mitral valve intervention for patients with severe aortic valve disease. ² Nevertheless, controversies exist in whether moderate FMR should be surgically intervened during SAVR procedure. ³ , ⁴ Many surgeons believe that moderate FMR would automatically improve after SAVR, therefore requiring no operations on the mitral valve. ⁵ , ⁶ However, several studies indicate that patients with deterioration of the FMR are not rare cases after isolated SAVR, which could potentially increase the long‐term mortality. ⁷ , ⁸ , ⁹ Therefore, treatments of aortic valve disease patients complicated with moderate FMR are an urgent clinical issue to be solved.

Increased haemodynamic load (including the preload and afterload) and neurohormonal activation are the critical risk factors of ventricular remodelling. ¹⁰ β‐Blockers (BB), a widely used medication in various cardiovascular diseases, have proved to be effective to reverse the remodelling of left ventricle and improve clinical outcomes in patients undergoing cardiothoracic surgeries. ¹¹ , ¹² As is the case with systolic heart failure, volume overload due to mitral regurgitation could result in neuroendocrine activation within the circulation, leading to an elevated β‐adrenergic state, decreased synthesis of myocyte protein, and reduced degradation of extracellular matrix. ¹³ Directly targeted at β‐adrenergic receptors, BB has been shown to effectively suppress cardiac remodelling through several classical signalling pathways. ¹⁰ Therefore, BB might be potentially beneficial for improving the prognosis of aortic valve disease patients undergoing SAVR but complicated with moderate FMR. So far, evidence is lacking regarding the impact of post‐operative administration of BB for this special group of patients. Therefore, the current study aimed to assess the prognostic impacts of BB in severe aortic valve disease patients complicated with moderate FMR after the SAVR.

Methods

Study design

In this retrospective single‐centred cohort study, we recruited patients who underwent isolated SAVR and complicated with moderate FMR pre‐operatively between January 2010 and December 2019 at our institute. The investigation conforms with the principles outlined in the Declaration of Helsinki, and the institutional review board approved the use of the clinical data for this study and waived individual informed consent. This work has been reported in line with the STROCSS criteria. ¹⁴

Patient selection and stratification

The inclusion criteria included (i) adult patients with severe aortic valve disease, (ii) complication with moderate FMR, and (iii) treated by isolated SAVR therapy. Excluded patients were those with primary mitral valve disease, a history of infective endocarditis, or those who were under the age of 18 years. Specifically, FMR was identified by the presence of mitral regurgitation without any evidence of primary lesion on the mitral valve leaflets and chords. FMR in patients of the current cohort were attributed to ventricular causes, as all the included patients presented with severe aortic valve disease in need of surgical treatment. Although myocardial ischaemia could result in ischaemic FMR, the primary disease for the current cohort was severe aortic valve disease, which was the dominant causes for FMR. In fact, only 17.0% patients had co‐existing ischaemic heart disease, which might have posed limited impacts for the progression of FMR. Degree of mitral regurgitation, which was determined by the vena contracta and regurgitant jet area, was evaluated using transthoracic echocardiography at least for twice pre‐operatively and was stratified into five groups (0⁺ = no, 1⁺ = trace, 2⁺ = mild, 3⁺ = moderate, 4⁺ = severe), and patients with 3⁺ level were enrolled. Patients were divided into BB group and control group, depending on whether they received post‐operative BB therapy or not. Patients in the BB group were mostly prescribed with atenolol (88.9%) therapy, whereas metoprolol (11.1%) was prescribed to the remaining.

The primary endpoint was the major adverse cardiovascular and cerebrovascular events (MACCE), which was defined as the composite endpoint of all‐cause death, repeat heart valve surgery, non‐fatal myocardial infarction, stroke, and hospitalization for heart failure. The secondary endpoints were the changes in the post‐operative and the follow‐up echocardiographic parameters, including the left ventricular ejection fraction (EF), left ventricular end‐diastolic diameter (LVEDD), left atrial diameter (LAD), and the status of FMR.

Results of post‐operative echocardiographic examinations before discharge were retrieved from the electronic medical record system, whereas follow‐up echocardiographic parameters were collected at 3, 6, and 12 months at the outpatient visits after the index hospitalization. Operative death was defined as death within 30 days, despite its occurrence before or after discharge. The improvement of FMR was defined by decrement of mitral regurgitation for one degree or above. Other baseline characteristics were obtained by reviewing the electronic medical records. Phone call interview was used for patients who were unavailable for outpatient follow‐up.

Statistical methods

Shapiro–Wilk test was used to confirm the normality of continuous variables. Continuous variables were presented as mean ± standard deviation (SD) if normally distributed and tested by Student's t‐test. Otherwise, medians with the 25th and 75th percentiles were presented and tested by Kruskal–Wallis H test. Categorical variables were presented as numbers (%) and tested by chi‐square test, or by Fisher exact test, as appropriate. The overall and MACCE‐free survival were calculated using the Kaplan–Meier method and tested by the log‐rank test. Inverse probability treatment weighting (IPTW) analysis was performed to balance the potential confounders, in order to increase the comparability of the two groups. Variables included in the IPTW analysis were as follows: pre‐operative characteristics including age, sex, body mass index, body surface area, history of atrial fibrillation, renal dysfunction, hypertension, dyslipidaemia, coronary artery disease, coronary artery bypass grafting, smoking, diabetes mellitus, stroke, pre‐operative EF, LVEDD, LAD, type of aortic valve disease (stenosis or insufficiency), New York Heart Association Class III or IV, and intra‐operative characteristics including the type of prostheses (mechanical or bioprosthetic), concomitant coronary artery bypass grafting, concomitant tricuspid valve repair, duration of cardiopulmonary bypass, cross‐clamp time, and prescription of angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker. Besides, more echocardiographic parameters, such as left ventricular wall thickness, aortic valve peak velocity, and transaortic pressure gradient, were also balanced using IPTW method in the subgroup analysis. A standardized mean difference (SMD) < 0.2 or P value > 0.05 was considered to indicate adequate balance for between‐group differences. A P value < 0.05 was considered statistically significant. Statistical analyses were performed using R 4.0.2 (R Core Team, Vienna, Austria).

Results

Baseline and operative characteristics

A total of 165 patients were enrolled in the current cohort. Of all the patients, 57 (34.6%) were female, and the mean age was 59.2 ± 12.2 years. All patients underwent isolated SAVR without surgical intervention of the mitral valve, and 80 (48.5%) patients received post‐operative BB therapy (Table 1 ). The BB group acquired higher prevalence for the history of coronary artery disease (23.8 vs. 10.6%, P = 0.024, SMD = 0.354) but lower prevalence of smoking history (36.2 vs. 51.8%, P = 0.045, SMD = 0.316). No significant differences were detected regarding the preoperative echocardiographic parameters between the two groups. The EF level was classified into ≤40%, 41–49%, and ≥50%. It is worth mentioning that the distributions of EF were also comparable between the two groups (control: 15.3, 14.1, and 70.6%, respectively; BB: 7.5, 22.5, and 70.0%, respectively, P = 0.152). As for the intraoperative profiles (Table 2 ), BB group possessed longer cross‐clamp time [68.0 (53.0, 90.0) vs. 73.5 (60.0, 109.0) min, P = 0.039, SMD = 0.296] and more frequently received concomitant coronary artery bypass grafting (9.4 vs. 23.8%, P = 0.013, SMD = 0.393).

Table 1.

Baseline characteristics of the overall cohort

Variables	Original				IPTW
	Control group (N = 85)	BB group (N = 80)	P value	SMD	Control group (N = 156.54)	BB group (N = 170.35)	P value	SMD
Age, years	60.0 ± 11.7	58.5 ± 12.9	0.510	0.103	59.6 ± 12.2	59.4 ± 11.7	0.924	0.015
Female, n (%)	25 (29.4)	32 (40.0)	0.153	0.224	50.9 (32.5)	55.0 (32.3)	0.978	0.005
Body mass index, kg/m2	23.5 [20.1, 26.3]	22.5 [19.9,25.6]	0.422	0.142	23.1 [19.0, 25.3]	22.5 [19.8, 25.6]	0.936	0.026
Body surface area, m2	1.8 [1.6, 1.9]	1.7 [1.6, 1.8]	0.249	0.170	1.8 [1.6, 1.9]	1.7 [1.6, 1.9]	0.802	0.024
Atrial fibrillation, n (%)	10 (11.8)	7 (8.8)	0.524	0.099	18.4 (11.7)	20.0 (11.7)	>0.99	<0.001
Hypertension, n (%)	33 (38.8)	31 (38.8)	0.992	0.002	60.3 (38.5)	61.9 (36.3)	0.780	0.046
Dyslipidaemia, n (%)	25 (29.4)	25 (31.2)	0.797	0.040	44.4 (28.4)	46.3 (27.2)	0.868	0.027
Coronary artery disease, n (%)	9 (10.6)	19 (23.8)	0.024 a	0.354	20.3 (13.0)	25.3 (14.9)	0.729	0.055
Prior CABG	0	1 (1.2)	0.301	0.159	0	1.0 (0.6)	0.345	0.109
Stroke, n (%)	8 (9.4)	7 (8.8)	0.883	0.023	13.5 (8.6)	12.9 (7.6)	0.793	0.040
Renal failure, n (%)	2 (2.4)	2 (2.5)	0.951	0.010	2.8 (1.8)	2.8 (1.6)	0.942	0.010
NYHA Class III or IV, n (%)	40 (47.1)	32 (40.0)	0.361	0.143	68.1 (43.5)	73.9 (43.4)	0.986	0.003
Diabetes mellitus, n (%)	12 (14.1)	10 (12.5)	0.760	0.048	20.8 (13.3)	26.2 (15.4)	0.733	0.060
Smoking, n (%)	44 (51.8)	29 (36.2)	0.045 a	0.316	74.2 (47.4)	80.7 (47.4)	>0.99	<0.001
Aortic valve disease, n (%)			0.985	0.003			0.941	0.012
Aortic regurgitation	49 (57.6)	46 (57.5)			89.8 (57.4)	98.8 (58.0)
Aortic stenosis	36 (42.4)	34 (42.5)			66.7 (42.6)	71.6 (42.0)
Pre‐operative
EF, %	55.0 [46.0, 61.0]	55.5 [46.8, 60.0]	0.914	0.027	55.0 [46.0, 61.0]	55.0 [45.0, 60.0]	0.845	0.018
LAD, mm	43.0 [38.0, 48.0]	40.5 [38.0, 46.3]	0.303	0.117	43.00 [38.0, 48.0]	42.0 [38.0, 47.0]	0.811	0.005
LVEDD, mm	63.0 [55.0, 71.0]	60.0 [52.8, 68.0]	0.132	0.243	61.0 [54.3, 70.1]	62.8 [53.4, 70.0]	0.979	0.008

BB, β‐blocker; CABG, coronary artery bypass grafting; EF, ejection fraction; IPTW, inverse probability treatment weighting; LAD, left atrial diameter; LVEDD, left ventricular end‐diastolic diameter; NYHA, New York Heart Association; SMD, standardized mean difference.

^a Statistically significant.

Table 2.

Intra‐operative and post‐operative characteristics of the overall cohort

Variables	Original				IPTW
	Control group (N = 85)	BB group (N = 80)	P value	SMD	Control group (N = 156.54)	BB group (N = 170.35)	P value	SMD
Operative characteristics
CPB duration, min	96.0 [74.0, 121.0]	114.5 [84.8, 142.3]	0.057	0.274	96.0 [73.7, 131.0]	96.3 [81.4, 128.5]	0.756	0.020
Cross‐clamp time, min	68.0 [53.0, 90.0]	73.5 [60.0, 109.0]	0.039	0.296	70.0 [53.0, 96.7]	69.3 [56.0, 96.4]	0.677	0.097
Mechanical valve, n (%)	55 (64.7)	51 (63.7)	0.898	0.020	99.6 (63.7)	103.8 (61.0)	0.744	0.056
Concomitant CABG, n (%)	8 (9.4)	19 (23.8)	0.013 a	0.393	19.3 (12.3)	25.3 (14.9)	0.644	0.074
Concomitant TV surgery, n (%)	3 (3.5)	3 (3.8)	0.940	0.012	5.7 (3.7)	5.6 (3.3)	0.896	0.020
Early post‐operative results
Perioperative transfusion, n (%)	14 (16.5)	12 (15.0)	0.796	0.040	28.1 (17.9)	22.6 (13.2)	0.422	0.130
New‐onset atrial fibrillation, n (%)	3 (3.5)	4 (5.0)	0.640	0.073	4.2 (2.7)	8.3 (4.9)	0.428	0.116
Acute kidney injury, n (%)	9 (10.6)	5 (6.2)	0.318	0.157	14.0 (8.9)	11.9 (7.0)	0.707	0.072
ACEI/ARB, n (%)	13 (15.3)	8 (10.0)	0.308	0.160	20.3 (13.0)	21.3 (12.5)	0.938	0.014
EF, %	55.0 [46.0, 60.0]	55.0 [47.5, 60.0]	0.647	0.099	55.0 [46.0, 60.0]	53.0 [44.8, 60.0]	0.606	0.075
LAD, mm	35.0 [32.0, 39.0]	35.0 [31.0, 38.0]	0.666	0.033	35.0 [32.0, 39.0]	36.0 [31.0, 38.3]	0.645	0.124
LVEDD, mm	53.0 [47.0, 60.0]	51.0 [46.8, 56.0]	0.161	0.275	51.6 [46.3, 59.3]	53.6 [47.0, 59.6]	0.972	0.052
Mitral regurgitation, n (%)			0.382	0.217			0.122	0.324
No	50 (58.8)	51 (63.7)			88.3 (56.4)	106.2 (62.3)
Mild	34 (40.0)	26 (32.5)			67.0 (42.8)	55.2 (32.4)
Moderate	1 (1.2)	3 (3.8)			1.2 (0.8)	9.0 (5.3)
Improved	84 (98.8)	77 (96.3)	0.283	0.167	155.3 (99.2)	161.3 (94.7)	0.055	0.265

ACEI, angiotensin‐converting enzyme inhibitor; ARB, angiotensin receptor blocker; BB, β‐blocker; CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass; EF, ejection fraction; IPTW, inverse probability treatment weighting; IQR, interquartile range; LAD, left atrial diameter; LVEDD, left ventricular end‐diastolic diameter; SMD, standardized mean difference; TV, tricuspid valve.

^a Statistically significant.

Early post‐operative results

No operative deaths occurred in the two groups, and no significant differences were observed regarding the incidence of post‐operative new‐onset atrial fibrillation (3.5 vs. 5.0%, P = 0.640, SMD = 0.073), acute kidney injury (10.6 vs. 6.2%, P = 0.318, SMD = 0.157), and the use of perioperative transfusion (16.5 vs. 15.0%, P = 0.796, SMD = 0.040). Patients in both groups showed similar rate of postoperative improvement of FMR (98.8 vs. 96.3%, P = 0.283, SMD = 0.167). Besides, no significant difference was observed for other echocardiographic parameters between the two groups, including LAD, LVEDD, and EF (Table 2 ).

Follow‐up outcomes

All patients in the cohort had completed the follow‐up for clinical outcomes, either by outpatient visit or phone call interview. During a median follow‐up of 18.4 [12.1–18.3] months, a total of 21 (12.7%) cases of MACCE were reported, including 10 (6.1%) cases of all‐cause death, 5 (3.0%) cases of stroke, 9 (5.5%) cases of rehospitalization due to heart failure, and one (0.6%) case of myocardial infarction (Table 3 ). Regarding the death cases, eight of them were cardiac death, whereas the other two were due to haemorrhagic stroke and malignancy. According to the Kaplan–Meier analysis, the MACCE‐free survival was similar for BB users and non‐users (Figure 1 ). The administration of BB was not associated with lower risk of MACCE [HR (hazard ratio): 0.68, 95% confidence interval (CI): 0.29–1.62, P = 0.388] or all‐cause mortality (HR: 1.03, 95% CI: 0.30–0.56, P = 0.967).

Table 3.

Follow‐up outcomes of the overall cohort

Variables	Original			IPTW
	Control group (N = 85)	BB group (N = 80)	P value	Control group (N = 156.54)	BB group (N = 170.35)	P value
All‐cause death, n (%)	5 (5.9)	5 (6.3)	0.967 a	8.4 (5.3)	12.1 (7.1)	0.674 a
Cardiac death	4	4		‐	‐
Haemorrhagic stroke	1	0		‐	‐
Lung cancer	0	1		‐	‐
Stroke, n (%)	4 (4.7)	1 (1.2)		7.8 (5.0)	3.9 (2.3)
Heart failure, n (%)	6 (7.1)	3 (3.8)		10.1 (6.5)	4.9 (2.9)
Myocardial infarction, n (%)	0	1 (1.2)		0	1.2 (0.7)
MACCE	12 (14.1)	9 (11.2)	0.388 a	21.7 (13.9)	20.5 (12.0)	0.607 a

BB, β‐blocker; IPTW, inverse probability treatment weighting; MACCE, major adverse cardiovascular and cerebrovascular events.

^a Log‐rank test.

Figure 1.

Comparison of overall (A,C) and MACCE‐free (B,D) survival. Panels A and B show the unmatched results, whereas Panels C and D show the results of IPTW analysis. BB, β‐blocker; IPTW, inverse probability treatment weighting. MACCE, major adverse cardiovascular and cerebrovascular events.

Regarding the follow‐up of cardiac function parameters, 119 patients (72.2%) had undergone re‐examinations of echocardiography. Most of the measurements were similar between patients with or without BB treatment, including EF [60.0 (55.0, 65.0) vs. 60.0 (55.0, 62.3) %, P = 0.312], LAD [37.0 (34.0, 40.0) vs. 36.00 (33.0, 39.3) mm, P = 0.433], LVEDD [49.0 (44.0, 54.0) vs. 48.0 (46.0, 53.0) mm, P = 0.767], and the occurrence of improvement in FMR (93.7 vs. 98.2%, P = 0.216). Besides, the decrease of LVEDD is substantially greater in the BB group when compared to post‐operative baseline measurements [BB vs. control group: −4.0 (−8.5, 0) vs. −1.5 (−5.3, 1.0) mm, P = 0.023]. However, changes of EF [5.0 (0, 12.0) vs. 5.0 (0, 10.0) %, P = 0.386] and LAD [2.0 (−1.0, 5.0) vs. 3.0 (−1.0, 6.0) mm, P = 0.411] were similar between groups (Table 4 ).

Table 4.

Follow‐up echocardiographic results of the overall cohort

Variables	Original			IPTW
	Control group (N = 63)	BB group (N = 58)	P value	Control group (N = 116.0)	BB group (N = 113.3)	P value
EF, %	60.0 [55.0, 65.0]	60.0 [55.0, 62.3]	0.312	60.0 [55.5, 65.0]	60.0 [55.0, 62.0]	0.163
ΔEF	5.0 [0, 12.0]	5.0 [0, 10.0]	0.386	5.0 [0, 11.0]	5.0 [−1.0, 10.0]	0.629
LAD, mm	37.0 [34.0, 40.0]	36.0 [33.0, 39.3]	0.433	36.9 [33.2, 40.0]	36.7 [33.0, 41.1]	0.886
ΔLAD	2.0 [−1.0, 5.0]	3.0 [−1.0, 6.0]	0.411	1.6 [−1.0, 5.0]	3.3 [−1.0, 6.0]	0.228
LVEDD, mm	49.0 [44.0, 54.0]	48.0 [46.0, 53.0]	0.767	47.0 [43.9, 53.2]	48.2 [47.0, 54.1]	0.190
ΔLVEDD	−4.0 [−8.5, 0]	−1.5 [−5.3, 1.0]	0.023 a	−3.6 [−8.5, 0]	−2.0 [−5.3, 1.0]	0.049 a
FMR, n (%)			0.081			0.003 a
No	45 (41.3)	33 (56.9)		81.6 (70.3)	66.6 (58.8)
Trivial or mild	14 (19.4)	22 (27.5)		25.4 (21.9)	44.8 (39.5)
Moderate	4 (6.4)	1 (1.7)		9.0 (7.8)	1.9 (1.7)
Improved	59 (93.7)	55 (98.2)	0.216	107.0 (92.2)	111.4 (98.3)	0.033 a

BB, β‐blocker; EF, ejection fraction; IPTW, inverse probability treatment weighting; LAD, left atrial diameter; LVEDD, left ventricular end‐diastolic diameter.

Δ Change in echocardiographic characteristics when compared with post‐operative parameters.

^a Statistically significant.

IPTW analysis

Because differences existed in several baseline and intra‐operative characteristics between the two groups, IPTW analysis was performed to balance these potential confounders in order to improve the between‐group comparability. In the IPTW dataset, the total number of patients were 326.89, and the critical baseline and intraoperative characteristics were well balanced, as all variables acquired SMD values <0.2 and P value > 0.05 (Tables 1 and 2 and Figure 2 ).

Figure 2.

Comparison of critical baseline and intra‐operative characteristics before (unmatched) and after (weighted) IPTW analysis. ACEI, angiotensin‐converting enzyme inhibitor; ARB, angiotensin receptor blocker; CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass; EF, ejection fraction; LAD, left atrial diameter; LVEDD, left ventricular end‐diastolic diameter; NYHA, New York Heart Association; SMD, standardized mean difference; TV, tricuspid valve.

Regarding the early post‐operative outcomes, the incidence of perioperative transfusion (17.9 vs. 13.2%, P = 0.422), new‐onset atrial fibrillation (2.7 vs. 4.9%, P = 0.428), and acute kidney failure (8.9 vs. 7.0%, P = 0.707) were comparable between patients with or without BB treatment. Similar results were observed for the improvement of moderate FMR, LAD, and LVEDD (Table 2 ).

Regarding the follow‐up outcomes (Table 3 ), the administration of BB was not associated with lower risk of MACCE (HR: 0.79, 95% CI: 0.31–1.97, P = 0.607) or all‐cause mortality (HR: 1.33, 95% CI:0.35–5.05, P = 0.674), which was in line with the unmatched analysis. Notably, more patients in the BB group acquired improvement of moderate FMR (92.2 vs. 98.3%, P = 0.033) but smaller decrease in LVEDD [−3.6 (−8.5, 0) vs. −2.0 (−5.3, 1.0) mm, P = 0.049] as compared with the control group (Table 4 ).

Subgroup analysis and sensitivity analysis

To furtherly investigate the prognostic effect of BB on the patient outcome, patients were stratified into two subgroups, namely, aortic stenosis and aortic insufficiency. Baseline, intra‐operative, and follow‐up outcomes were compared (Table S1–S3 ). The use of BB was not associated with lower risk of MACCE in both the aortic stenosis (HR: 0.44, 95% CI: 0.07–2.67, P = 0.373) and aortic insufficiency (HR: 0.87, 95% CI: 0.32–2.35, P = 0.789) patients, which was sustained in the IPTW analysis. In addition, more patients in the BB group showed better improvement of moderate FMR in the aortic stenosis subgroup (96.2 vs. 100%, P > 0.99) and aortic insufficiency subgroup (91.9 vs. 97.0%, P = 0.616) during the follow‐up, even though it failed to reach the statistical significance. In the sensitivity analysis, patients without coronary artery disease were included, and the results were similar to those in the subgroup analysis (Table S4 and S5 ).

Discussion

In this study, we evaluated the impact of post‐operative administration of BB on the outcomes of patients with severe aortic valve disease combined with moderate FMR treated by SAVR, for which the major findings were as follows: (i) BB did not improve the MACCE‐free or overall survival during the mid‐term follow‐up, as shown by the unmatched and IPTW analysis; (ii) the use of BB was associated with the improvement of moderate FMR, but not the reduction of LVEDD.

Controversies for the treatment of FMR in severe aortic valve diseases

The enlargement of left ventricle plays an important role in the development of FMR among patients with severe aortic valve disease, including aortic stenosis and insufficiency. It is well known that aortic valve disease could result in the hypertrophy and dilation of left ventricle due to the increased preload or afterload, causing the dilation of mitral annulus and incomplete coaptation of the mitral leaflets, which finally leads to the incidence of FMR. Therefore, SAVR has been one of the major treatments for these patients, because this procedure could resolve the abnormal haemodynamics caused by severe aortic valve disease, thereby limiting the progression of left ventricular remodelling and functional valvular regurgitation.

Despite the controversies to amend mitral valve during the valvular replacement procedure, ¹ , ³ , ⁴ , ⁹ , ¹⁵ many surgeons believe that an isolated SAVR is more reasonable, ⁵ , ⁶ because the preload and afterload could be substantially reduced following the anatomical reconstruction of aortic valve after SAVR. Theoretically, the ongoing left ventricular remodelling and mitral valve regurgitation would also be partially suppressed, therefore attenuating the clinical significance for the correction of FMR. However, real‐world studies suggest that the efficacy for SAVR to resolve FMR is questionable, especially for patients complicated with certain risk factors, ¹⁶ leading to increased mortality for deferring the treatment of FMR. ⁷ , ⁸ , ⁹ Therefore, effective treatment is needed to relieve FMR in patients with severe aortic valve disease, with the expectation to inhibit ventricular remodelling, preserve cardiac function, and improve patient outcomes after SAVR.

Prognostic impacts of β‐blockers on FMR

As a classical medication in the field of cardiology, BB has been proved effective for improving clinical outcomes and cardiac function in patients with a wide range of cardiovascular diseases, ¹⁷ , ¹⁸ , ¹⁹  such as heart failure, myocardial infarction, and cardiomyopathy. Recent study even shows that the use of BB is associated with better outcomes in patients undergoing cardiac surgery. ¹¹ , ¹² However, debates still exist for the usage of BB in patients with valvular heart disease. ²⁰ Although a large number of patients complicated with severe aortic valve disease receive BB therapy after the surgery in the clinical practice, researchers report that these medications might increase the mortality in patients with severe chronic non‐ischaemic mitral regurgitation. ¹³ A recent study by Schubert et al. even shows that that pre‐operative BB treatment is associated with higher risk of post‐operative adverse events in patients undergoing surgical SAVR. ²¹ On the contrary, Saito et al. report that the use of BB has improved the survival of patients receiving transcatheter SAVR. ¹¹ Nevertheless, there is no study on the impact of BB on the post‐operative outcomes in patients with severe aortic valve disease complicated by moderate FMR after SAVR. For this specific group of patients, results from the current study suggest that the administration of BB did not bring extra clinical benefit in the early and mid‐term event‐free survival, even in the subgroup analysis and sensitivity analysis, which was consistent with the prior studies.

Abundant evidence from basic research suggests that β‐adrenergic regulation plays critical role in the process of cardiac remodelling. ¹⁰ Many clinical studies also show that BB could effectively suppress the progression of left ventricular remodelling, ²² , ²³ which directly contributes to the improvement of left ventricular function. ¹⁷ Sahoo et al. have also reported that the usage of BB is able to improve left ventricular parameters in a group of patients with rheumatic mitral regurgitation. ²⁴ Interestingly, results in the current study suggested that BB was associated with the higher chances for the improvement of moderate FMR, but smaller reduction of LVEDD. Currently, it is generally confirmed that BB reduce the mortality of patients of heart failure with reduced EF, ²⁵ , ²⁶ whereas controversies still exist regarding its efficacy for those with preserved EF. ²⁷ In a recent meta‐analysis including 3087 patients from 10 randomized trials, ²⁸ Marti et al. suggest that evidence for routine use of BB in patients with preserved cardiac function is limited, although the pooled analysis shows a possible 25% reduction of cardiovascular mortality in BB users, as the certainty of evidence is low. Barry et al. also reports in their review that BB has barely affects the outcomes of patients with aortic insufficiency and normal EF. ²⁰ Notably, however, we have included patients with various levels of EF regardless of their classifications of heart failure in this study, whereas the majority of patients acquired preserved or even numerically normal EF. The characteristics of the patient cohort might partially explain the neutral results observed in this study, as the clinical benefit of BB could have been attenuated where most of the patients possess normal cardiac function.

Limitations

This study has several limitations. Firstly, this was a retrospective cohort study, the data of which was from a single centre. Hence, the bias caused by the study design could not be avoided. Secondly, our study was also limited with relatively smaller sample size and shorter follow‐up time, and some patients underwent associated procedures, which might have compromised the statistical power of the test. For example, survival curve showed the trend that BB might impact the event‐free survival after 3 years. However, low number of patients was at risk in the tail of the curve, which might have dampened the analysis. In addition, as we mentioned elsewhere, most of the patients in our study were belonged to the heart failure with preserved or even normal EF, causing no chance to perform subgroup analysis stratified by the level of EF. Meanwhile, even though the IPTW analysis have its priority on balancing the baseline characteristics of the patients, unmeasured confounders could still be present. What is more, limited parameters of the left ventricle were available in this study, and not all of the patients completed the follow‐up echocardiography, which might also compromise the representativeness of the population to some extent. Last but by no means the least, although the patients were prescribed with BB, the medical compliance of the patients was not fully understood, which might also have caused bias in this study.

Conclusions

The administration of BB after SAVR was not associated with lower risk of MACCE for patients of severe aortic valve disease and moderate FMR, but was potentially beneficial for improving FMR.

Conflict of interest

All authors declare no conflict of interest.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not‐for‐profit sectors.

Supporting information

Table S1. Characteristics of the patients with aortic stenosis.

Table S2. Characteristics of the patients with aortic insufficiency.

Table S3. Follow‐up echocardiographic results of patients in the subgroup analysis.

Table S4. Sensitivity analysis including the patients without coronary artery disease.

Table S5. Follow‐up echocardiographic results of patients without coronary artery disease.

Tiemuerniyazi, X. , Nan, Y. , Song, Y. , Yang, Z. , Zhao, W. , Xu, F. , and Feng, W. (2022) Effect of β‐blocker on patients with moderate functional mitral regurgitation undergoing surgical aortic valve replacement. ESC Heart Failure, 9: 3317–3326. 10.1002/ehf2.14053.