Comparison of outcomes following valve-sparing aortic root replacement in patients with bicuspid and tricuspid aortic valves: A meta-analysis

Abstract

BACKGROUND

Aortic root dilation, linked to bicuspid aortic valve (BAV) or tricuspid aortic valve (TAV), risks aneurysm and dissection. Valve-sparing aortic root replacement (VSARR) preserves native valves, avoiding prosthetic valve complications. Long-term VSARR durability, especially in BAV patients, is debated. We hypothesize that VSARR outcomes differ between BAV and TAV patients in short-term and long-term settings.

AIM

To investigate short-term and long-term outcomes of VSARR in BAV vs TAV patients.

METHODS

This Preferred Reporting Items for Systematic Reviews and Meta-Analyses-compliant meta-analysis included observational studies comparing VSARR in adult BAV vs TAV patients. PubMed, ScienceDirect, and EMBASE were searched from inception to June 2025. Outcomes included mortality, reintervention, and procedural times. Pooled relative risk (RR) and mean differences (MD) with 95%CI were calculated. Risk of bias was assessed using Risk of Bias in Non-randomized Studies of Interventions; evidence certainty via GRADE.

RESULTS

Thirteen observational studies involving 1419 BAV and 2349 TAV patients were included. In-hospital mortality (RR = 0.34, 95%CI: 0.10–1.14, P = 0.08) and reoperation (RR = 1.04, 95%CI: 0.64–1.69, P = 0.87) showed no significant differences. All-cause mortality risk was significantly lower in BAV patients (RR = 0.34, 95%CI: 0.13–0.86, P = 0.02). Overall reintervention risk was significantly greater in BAV patients (RR = 2.64, 95%CI: 1.96–3.55, P < 0.00001). Aortic cross-clamp (MD = 3.35 minutes, 95%CI: -5.06 to 11.76, P = 0.43) and cardiopulmonary bypass times (MD = 3.96 minutes, 95%CI: -10.26 to 18.18, P = 0.59) showed no significant differences but substantial heterogeneity. The certainty of evidence was moderate for reintervention, low for mortality risk and in-hospital reoperation, and very low for procedural times.

CONCLUSION

VSARR demonstrates comparable short-term safety between BAV and TAV patients. However, BAV patients face a significantly higher long-term reintervention risk, highlighting the need for tailored strategies and further research.

Key Words: Valve-sparing aortic root replacement; Bicuspid aortic valve; Tricuspid aortic valve; Reintervention; Mortality

Core Tip: Valve-sparing aortic root replacement (VSARR) preserves native valves in patients with bicuspid aortic valve (BAV) or tricuspid aortic valve (TAV), reducing prosthetic valve complications. Our meta-analysis reveals comparable short-term safety between BAV and TAV patients but a significantly higher long-term reintervention risk in BAV patients. These findings highlight the need for tailored surgical strategies and vigilant long-term monitoring to optimize outcomes in BAV patients undergoing VSARR.

INTRODUCTION

Aortic root dilation, a critical condition often linked to bicuspid aortic valve (BAV) or tricuspid aortic valve (TAV), increases the risk of life-threatening complications such as aortic aneurysm and dissection[1,2]. Surgical intervention is frequently required to prevent catastrophic outcomes, with valve-sparing aortic root replacement (VSARR) gaining prominence as a technique that preserves the native aortic valve[3]. Unlike traditional composite valve conduit procedures, such as the Bentall procedure, VSARR eliminates the need for mechanical or bioprosthetic valves, thereby reducing the risks of anticoagulation-related complications, thromboembolism, and valve deterioration[4,5]. Developed through pioneering techniques such as reimplantation and remodeling by surgeons like Dr. Tirone David and Dr. Magdi Yacoub, VSARR has shown promising mid-term and long-term outcomes, making it an attractive option for patients with morphologically preserved aortic valves[6].

The choice of VSARR is particularly relevant for patients with BAV, a common congenital anomaly, and TAV, as both groups present unique anatomical and physiological challenges[7]. While VSARR has been widely adopted, its application in BAV patients remains debated due to concerns about long-term valve durability and the need for reintervention[8]. The growing body of clinical research underscores the importance of comparing outcomes between BAV and TAV patients to refine surgical strategies and improve patient prognosis.

Current VSARR techniques, including the David reimplantation and Yacoub remodeling procedures, aim to stabilize the aortic root while preserving valve function[6]. These approaches have demonstrated comparable short-term outcomes, with low in-hospital mortality and reoperation risk for both BAV and TAV patients[9]. However, challenges persist, particularly in BAV patients, who often require additional leaflet repairs due to cusp prolapse or commissural fusion, which can potentially increase the risk of recurrent aortic regurgitation (AR)[10]. Long-term data suggest that BAV patients may face higher reintervention rates, driven by factors such as aortic stenosis, endocarditis, or progressive valve dysfunction[8]. Additionally, variations in surgical expertise, patient selection, and the use of adjunctive techniques, such as aortic annuloplasty, complicate outcome comparisons, highlighting the need for standardized approaches and robust evidence.

Despite advances in VSARR, the comparative long-term outcomes for BAV vs TAV patients remain incompletely understood. A meta-analysis by Zuo et al[9] provided critical insights, indicating similar perioperative outcomes but a higher long-term reintervention rate in BAV patients. Emerging studies have provided new data on patient characteristics, surgical techniques, and long-term outcomes for VSARR in patients with BAV and TAV. However, these findings have not been systematically synthesized to address unresolved questions about valve durability, reintervention risks, and factors influencing differential outcomes. An updated and comprehensive analysis is needed to clarify these uncertainties and inform clinical practice. This meta-analysis aimed to evaluate the short-term and long-term outcomes of VSARR in BAV compared to TAV patients, integrating recent studies to build a robust evidence base. We seek to identify factors contributing to outcome differences and enhance surgical decision-making for patients with aortic root dilation.

MATERIALS AND METHODS

The review adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines[11] and is registered with PROSPERO (No. CRD420251088958).

Search strategy

We systematically searched PubMed, Cochrane Central Register of Controlled Trials, and EMBASE databases to identify studies comparing outcomes of VSARR in patients with BAV and TAV. The search included all studies published up to June 2025, without language or publication date restrictions. Search terms (“Aortic Valve”[MeSH Terms] OR “valve-sparing aortic root replacement”[Title/Abstract] OR “VSARR”[Title/Abstract] OR “aortic root replacement”[Title/Abstract] OR “David procedure”[Title/Abstract]) AND (“Bicuspid”[Title/Abstract] OR “Bicuspid Aortic Valve”[Title/Abstract] OR “BAV”[Title/Abstract]) AND (“Tricuspid Aortic Valve”[Title/Abstract] OR “TAV”[Title/Abstract]). Additional studies were identified by manually reviewing reference lists of included articles and relevant reviews. Detailed Search String for each database is given in the Supplementary material.

Eligibility criteria

Studies were included if they were observational studies or randomized controlled trials comparing outcomes of VSARR in adult patients (aged ≥ 18 years) with BAV vs TAV, reporting at least one of the following outcomes: (1) In-hospital mortality; (2) Reoperation rate; (3) Long-term mortality; and (4) Freedom from reintervention. Eligible studies involved VSARR techniques, such as reimplantation or remodeling, with or without adjunctive procedures like leaflet repair, and provided quantitative data suitable for meta-analysis, with no restrictions on language or publication date.

Studies were excluded if they were single-arm studies focusing solely on BAV or TAV, non-VSARR interventions, case reports, case series with fewer than 10 patients, editorials, reviews, or studies lacking comparative data or clear outcome reporting for both BAV and TAV groups. Pediatric studies, non-human studies, and duplicate or non-peer-reviewed publications were also excluded.

Study selection

Studies were selected through a systematic screening process based on predefined inclusion and exclusion criteria. Two independent reviewers (Adugna LF and Asfeha NF) assessed the titles and abstracts of all retrieved studies to identify those meeting the eligibility criteria. Full-text articles of potentially relevant studies were then reviewed to confirm their inclusion. Any discrepancies between the reviewers were resolved through discussion and consensus among the research team. This rigorous selection process ensured the inclusion of only high-quality, relevant studies in the meta-analysis.

Extraction

Two reviewers (Hailu SZ and Khan ZH) independently extracted data using a standardized form, including study details (author, country, study period, sample size, follow-up duration), patient characteristics [age, male gender, preoperative AR grades, ejection fraction (EF), Marfan syndrome, aortic dissection, prolapse correction], and outcomes. Primary outcomes were in-hospital mortality, in-hospital reoperation, overall mortality, and reintervention rates (overall; short-term: < 1 year; mid-term: 1–5 years; long-term: > 5 years). Secondary outcomes included aortic cross-clamp time (ACX) and cardiopulmonary bypass (CPB) time. Data were collected for the BAV and TAV groups. Discrepancies were resolved by consensus.

Quality assessment

The risk of bias assessment for the included non-randomized studies was conducted using the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) tool, which evaluates potential biases across seven domains: (1) Bias due to confounding; (2) Bias in the selection of participants; (3) Bias in classification of interventions; (4) Bias due to deviations from intended interventions; (5) Bias due to missing data; (6) Bias in measurement of outcomes; and (7) Bias in the selection of the reported result. Each study was independently assessed by two reviewers, with discrepancies resolved through discussion or consultation with a third reviewer. The risk of bias for each domain and overall was categorized as low, moderate, serious, critical, or no information[12].

Statistical analysis

Statistical analyses were conducted using Review Manager version 5.4. Dichotomous outcomes, such as mortality and reoperation rates, were analyzed using relative risk (RR) with 95%CI. Continuous outcomes, including ACX and CPB time, were evaluated through mean differences (MD). Forest plots were created for each outcome to facilitate visual data interpretation and funnel plots were generated to assess potential publication bias. A P value less than 0.05 was considered indicative of statistical significance. Sensitivity analyses, including leave-one-out methods, were conducted to assess the impact of individual studies on heterogeneity and model selection. Subsequently, the quality of evidence for each outcome was appraised using the GRADE approach, which considers factors such as risk of bias, inconsistency, imprecision, indirectness, and effect size to determine the overall certainty of the evidence.

RESULTS

Included studies

The systematic search of PubMed, ScienceDirect, and EMBASE, conducted up to July 2025, yielded 802 records. After removing 79 duplicates, 723 records underwent title and abstract screening, of which 614 were excluded due to irrelevance, non-human studies, or inappropriate study designs. The full-text review was performed for 109 articles, with 96 excluded for lack of comparative BAV vs TAV data, insufficient outcome reporting, single-arm studies, and duplicate or overlapping data. Ultimately, 13 studies were included in the meta-analysis (Figure 1)[13-25].

Figure 1 Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart outlining the literature screening process, study selection, and exclusion criteria.

Study characteristics

The meta-analysis included 13 studies conducted across seven countries (Germany, Spain, Lithuania, the United States, Canada, Italy, and Belgium) from 1988 to 2023, with follow-up durations ranging from 4.1 years to 15 years. A total of 1419 patients with BAV and 2349 with TAV underwent VSARR, with sample sizes ranging from 24 to 414 for BAV and ranging from 63 to 589 for TAV groups. Mean or median age, reported in 10 studies, ranged from 40 years to 54 years for BAV patients and ranged from 36 years to 62 years for TAV patients, with BAV patients generally younger. Male gender, reported in 10 studies, predominated in both groups, with 77.2%–100% of BAV patients and 67.6%–93.1% of TAV patients being male. Three studies (Schäfers et al[14], 2015, Miyahara et al[15], 2020, Malvindi et al[16], 2012) did not report age or gender data. Detailed study characteristics are presented in Table 1[13-25].

Table 1 Characteristics of included studies, n (%).

Ref.	Country	Study period	Follow-up (years)	Sample size (n)		Age (years)		Male gender
				BAV	TAV	BAV	TAV	BAV	TAV
Aicher et al[25], 2007	Germany	1995–2006	10	81	193	52 ± 12	62 ± 15	69 (85.2)	132 (68.4)
Martín et al[19], 2017	Spain	2004–2015	5	57	103	46.0 ± 11.8	57.5 ± 17.8	57 (100)	89 (88.1)
Karciauskas et al[21], 2019	Lithuania	2004–2016	10	29	63	42.4 ± 12	55.3 ± 14.9	56 (88.9)	27 (93.1)
Schäfers et al[14], 2015	Spain	1995–2013	15	290	431	54 ± 15	–	–	–
Kvitting et al[17], 2013	United States	1993–2009	10	63	170	43 ± 12	36 ± 13	50 (79.4)	115 (67.6)
Bavaria et al[18], 2015	United States	2004–2013	5	40	89	46 ± 12	45 ± 15	35 (87.5)	63 (70.8)
Shrestha et al[20], 2018	Germany	1993–2015	8.7	24	173	40 (30–47)	49 (35–62)	21 (87.5)	123 (71.1)
Ouzounian et al[22], 2019	Canada	1988–2012	8.2	45	135	40 ± 13	41 ± 14	121 (89.6)	39 (86.7)
Malvindi et al[16], 2012	Italy	2002–2011	5	24	108	–	–	–	–
Miyahara et al[15], 2020	Germany	1995–2018	15	414	589	–	–	–	–
Mokashi et al[13], 2022	United States	2002–2017	8	71	71	48 ± 12	49 ± 12	57 (80.3)	52 (73.2)
François et al[24], 2025	Belgium	2000-2023	7.9	65	171	41.9 ± 14.4	48.7 ± 18.2	51 (77.2)	129 (74.6)
Levine et al[23], 2023	United States	2005-2020	4.1	156	156	45.00 ± 12.72	46.00 ± 12.72	131 (84)	133 (85.3)

BAV: Bicuspid aortic valve; TAV: Tricuspid aortic valve.

Patient characteristics

The meta-analysis included 13 studies with a total of 1419 BAV and 2349 TAV patients undergoing VSARR, with sample sizes ranging from 24 to 414 for BAV and ranging from 63 to 589 for TAV groups. Preoperative AR data showed a mix of severity grades, with grade I AR reported in 19–224 patients and grade III AR in 20–436 patients across both groups, though reporting was inconsistent. EF, reported in eight studies, ranged from 53.6% ± 7.5% to 88% in BAV patients and ranged from 48.7% ± 10.1% to EF ≥ 60% in TAV patients, indicating generally preserved ventricular function. Marfan syndrome was more prevalent in TAV patients (1–91 cases) than BAV patients (0–60 cases) across seven studies reporting this variable. Aortic dissection was less common, with 0–29 cases in BAV and 0–59 cases in TAV groups (five studies). Prolapse correction was frequently performed, particularly in BAV patients (10–70 cases) compared to TAV patients (3–103 cases) in eight studies. Variability in reporting limited pooled estimates for some characteristics, with detailed data presented in Table 2[13-25].

Table 2 Patient characteristics of the included studies.

Ref.	Sample size (n)		Aortic regurgitation grade I				EF (%)		Marfan syndrome		Dissection		Prolapse correction
	BAV	TAV	Grade I	Grade II	Grade III	Grade IV	BAV	TAV	BAV	TAV	BAV	TAV	BAV	TAV
Aicher et al[25], 2007	81	193	71	64	116	23	–	–	0	5	6	40	70	103
Martín et al[19], 2017	57	103	224	37	3	3	85.40	≥ 55	107	7	–	–	47	56
Karciauskas et al[21], 2019	29	63	5	47	38	2	53.6 ± 7.5	48.7 ± 10.1	–	–	1	6	28	53
Schäfers et al[14], 2015	290	431	56	183	436	21	–	–	–	–	29	59	–	–
Kvitting et al[17], 2013	63	170	89	64	52	28	62	61	3	91	0	0	42	63
Bavaria et al[18], 2015	40	89	19	38	20	28	58 ± 9	56 ± 10	0	47	–	–	40	13
Shrestha et al[20], 2018	24	173	21	29	50	97	–	–	60	–	–	–	11	14
Ouzounian et al[22], 2019	45	135	–	–	–	–	75	≥ 60	4	71	–	–	42	59
Malvindi et al[16], 2012	24	108	–	–	–	–	88	> 45	5	1	–	–	10	3
Miyahara et al[15], 2020	414	589	–	–	–	–	–	–	–	–	–	–	–	–
Mokashi et al[13], 2022	71	71	58	33	29	22	57 ± 5.4	57 ± 5.6	–	–	0	1	–	–
François et al[24], 2025	65	171	–	–	–	–	–	–	5 (7.6)	64 (36.9)	4 (6.1)	26 (15)	–	–
Levine et al[23], 2023	156	156	–	–	–	–	56.33 ± 4.49	56.67 ± 3.74	–	–	–	–	–	–

BAV: Bicuspid aortic valve; EF: Ejection fraction; TAV: Tricuspid aortic valve.

Clinical outcomes

In-hospital mortality: Five studies reported in-hospital mortality. The pooled risk ratio was 0.34 (95%CI: 0.10–1.14, P = 0.08), indicating no significant difference in mortality risk between BAV and TAV patients. Heterogeneity was low (I² = 0%) (Figure 2A)[19,21,22,24,25].

Figure 2 Forest plots of in-hospital mortality, all-cause mortality, and in-hospital reoperation comparing outcomes of valve-sparing aortic root replacement in bicuspid aortic valve vs tricuspid aortic valve patients. A: In-hospital mortality; B: All-cause mortality; C: In-hospital reoperation. BAV: Bicuspid aortic valve; TAV: Tricuspid aortic valve.

All-cause mortality: Four studies reported overall mortality. The pooled RR was 0.34 (95%CI: 0.13–0.86, P = 0.02), showing a significantly lower mortality risk in BAV patients compared to TAV patients. Heterogeneity was low (I² = 0%) (Figure 2B)[13,20,21,25].

In-hospital reoperation: Seven studies reported in-hospital reoperation. The pooled RR was 1.04 (95%CI: 0.64–1.69, P = 0.87), suggesting similar reoperation risks for BAV and TAV patients. Heterogeneity was low (I² = 18%) (Figure 2C)[13,18-23].

Reintervention

Overall reintervention: Eleven studies reported reintervention events. The pooled RR was 2.64 (95%CI: 1.96–3.55, P < 0.00001), indicating a significantly greater reintervention risk in BAV patients compared to TAV patients. Heterogeneity was low (I² = 22%) (Figure 3A)[13-16,19-25].

Figure 3 Forest plots of overall reintervention, short-term reintervention, mid-term reintervention, long-term reintervention comparing outcomes of valve-sparing aortic root replacement in bicuspid aortic valve vs tricuspid aortic valve patients. A: Overall reintervention; B: Short-term reintervention; C: Mid-term reintervention; D: Long-term reintervention. BAV: Bicuspid aortic valve; TAV: Tricuspid aortic valve.

Short-term reintervention: Eight studies reported short-term reintervention. The pooled RR was 2.75 (95%CI: 1.64–4.62, P = 0.0001), showing a significantly increased mortality risk in BAV patients compared to TAV patients. Heterogeneity was low (I² = 0%) (Figure 3B)[13,14,16,17,19-21,25].

Mid-term reintervention: Eight studies reported mid-term reintervention. The pooled RR was 2.32 (95%CI: 1.45–3.71, P = 0.0004), suggesting a significantly elevated mortality risk in BAV patients compared to TAV patients. Heterogeneity was low (I² = 0%) (Figure 3C)[13,14,16,17,19-21,25].

Long-term reintervention: Nine studies reported long-term reintervention. The pooled RR was 2.50 (95%CI: 1.82–3.43, P < 0.00001), indicating a significantly higher mortality risk in BAV patients compared to TAV patients. Heterogeneity was low (I² = 21%) (Figure 3D)[14-17,20-23,25].

Secondary outcomes

ACX time: Nine studies reported ACX time. The pooled MD was 3.35 minutes (95%CI: -5.06 to 11.76, P = 0.43), suggesting no significant difference in clamp time between BAV and TAV patients. Heterogeneity was substantial (I² = 93%) that dropped significantly upon exclusion of the Kvitting et al[17], 2013 and Aicher et al[25], 2007 (I² = 72%) (Figure 4A)[17-25].

Figure 4 Forest plots of aortic cross-clamp time, cardiopulmonary bypass time. A: Aortic cross-clamp time forest plot comparing procedural time outcomes of valve-sparing aortic root replacement in bicuspid aortic valve (BAV) vs tricuspid aortic valve (TAV) patients; B: Cardiopulmonary bypass time forest plot comparing outcomes of valve-sparing aortic root replacement in BAV vs TAV patients. BAV: Bicuspid aortic valve; TAV: Tricuspid aortic valve.

CPB time: Nine studies reported CBP time. The pooled MD was 3.96 minutes (95%CI: -10.26 to 18.18, P = 0.59), indicating comparable bypass times for BAV and TAV patients. Heterogeneity was substantial (I² = 94%), and dropped significantly upon exclusion of the Kvitting et al[17], 2013 and Aicher et al[25], 2007 (I² = 79%) studies (Figure 4B)[17-25]. Effect sizes for all clinical outcomes in BAV vs TAV patients are summarized in Supplementary Table 1.

Quality assessment

The ROBINS-I quality assessment revealed that the 13 included studies generally exhibited low-to-moderate risk of bias, supporting the reliability of the meta-analysis findings on VSARR outcomes in BAV and TAV patients. Five studies were rated as low risk overall, indicating robust study design and minimal bias, while six had moderate risk, primarily due to potential confounding or incomplete outcome reporting. Two studies, both older with smaller cohorts, were rated as serious risk, largely driven by missing data, which may limit their contribution to long-term outcome estimates. Detailed assessments are visualized in Supplementary Figures 1 and 2[13-25].

GRADE assessment

A total of nine outcomes were assessed across multiple observational studies, with the number of studies ranging from 4 to 11 for each outcome. While the risk of bias was not serious for most outcomes, one outcome (all-cause mortality) was downgraded due to a serious risk of bias in a contributing study.

Outcomes such as overall, short-term, mid-term, and long-term reintervention demonstrated moderate certainty of evidence, consistently indicating a significantly increased risk. Conversely, in-hospital mortality, all-cause mortality, and in-hospital reoperation showed low certainty of the evidence, primarily due to serious imprecision (wide confidence intervals crossing the null effect) or identified risk of bias. Finally, procedural times, specifically ACX time and CPB time, were associated with very low certainty of evidence, largely driven by substantial heterogeneity and serious imprecision, suggesting no significant difference between groups (Supplementary Table 2).

Publication bias

Publication bias was assessed using funnel plots for all outcomes. No evidence of publication bias was observed for in-hospital mortality, all-cause mortality, in-hospital reoperation, overall reintervention, short-term reintervention, mid-term reintervention, and long-term reintervention. ACX time and CPB time showed publication bias, indicated by asymmetrical funnel plots with gaps in the lower left quadrant, suggesting potential underrepresentation of smaller studies with non-significant or negative results (Supplementary Figure 3).

DISCUSSION

This meta-analysis provides comprehensive evidence regarding the comparative outcomes of VSARR between patients with BAV and TAV, revealing important insights that should inform clinical decision-making and patient counseling. The findings demonstrate comparable in-hospital mortality risk between BAV and TAV patients undergoing VSARR, indicating similar short-term safety profiles[9]. However, a significantly lower long-term mortality risk was observed in BAV patients, likely reflecting their younger age profile at the time of surgery[9]. The most clinically significant finding is the substantially higher reintervention risk in BAV patients, which persisted across short-term, mid-term, and long-term intervals. This increased reintervention rate can be attributed to several anatomical and pathophysiological factors unique to bicuspid valves, including asymmetric leaflet configuration, altered commissural anatomy, and frequent requirement for concomitant cusp repair[26,27]. The mechanism of failure also differs between valve types, with reintervention in BAV patients more likely due to progressive calcification and aortic stenosis, while TAV patients typically require reintervention for aortic insufficiency[26]. BAV patients often require additional procedures such as cusp plication (34% of cases) and commissure repair (8% of cases), which may contribute to the increased long-term failure rate[27].

The analysis found comparable procedural complexity between BAV and TAV patients, as evidenced by similar ACX and CPB times, although substantial heterogeneity was observed for these outcomes. This finding contrasts with historical concerns about increased technical difficulty in BAV repair, suggesting that modern surgical techniques and standardized repair protocols have helped standardize the approach to bicuspid valve preservation[28,29]. The comparable operative times indicate that experienced centers can achieve similar technical efficiency regardless of valve morphology, though the significant heterogeneity suggests variation in surgical techniques and institutional practices across centers. These findings have direct implications for patient selection and preoperative counseling, as the significantly higher reintervention risk in BAV patients must be carefully discussed with patients and families, particularly for younger BAV patients who have longer life expectancies and may face multiple reinterventions over their lifetimes[9,26,27].

The decision-making process should incorporate several factors beyond valve morphology, including patient age, life expectancy, and individual risk tolerance for reintervention[30,31]. For very young BAV patients (< 40 years), the higher reintervention risk may still be acceptable due to the decades of anticoagulation avoidance provided by valve preservation[30,32]. The results emphasize the critical importance of surgical expertise and center volume in achieving optimal outcomes with VSARR procedures, as studies have consistently demonstrated that the best results are achieved at high-volume centers with dedicated aortic valve specialists[33,34]. The United States National Marfan Foundation has established benchmarks requiring operative mortality < 1% and 10-year freedom from valve reoperation > 90% for centers performing these procedures, and this meta-analysis supports these recommendations, particularly for BAV patients, where the technical challenges are greater[35]. Centers should carefully evaluate their outcomes and consider referring complex BAV cases to specialized aortic centers if their reintervention rates exceed acceptable thresholds[30,33].

The higher reintervention risk in BAV patients necessitates enhanced long-term surveillance protocols, with current guidelines recommending annual echocardiographic evaluation; however, the timing and intensity of follow-up may need to be individualized based on valve morphology[26,34]. BAV patients may benefit from more frequent imaging, particularly beyond the 5-year mark when their reintervention risk appears to accelerate[32]. Recent advances in cardiac imaging, including four-dimensional (4D) flow magnetic resonance imaging (MRI) and advanced echocardiographic techniques, may help identify early signs of valve deterioration before clinical symptoms develop. The findings of this analysis are largely consistent with previous systematic reviews and meta-analyses examining VSARR outcomes, including a recent comprehensive meta-analysis by Zuo et al[9], similarly reporting no significant differences in early mortality risk but identifying trends toward higher reintervention rates in BAV patients. However, the current analysis provides more robust evidence with larger patient numbers and longer follow-up periods, and the observed reintervention rates align closely with large single-center series[24,25].

The patterns observed in this VSARR meta-analysis differ from those reported in surgical aortic valve replacement literature, where studies of prosthetic valve replacement generally show comparable long-term outcomes between BAV and TAV patients when modern prostheses are used[36,37]. This suggests that the anatomical complexity of bicuspid valves poses greater challenges for repair techniques than for replacement procedures, highlighting the importance of individualized treatment algorithms that consider not only valve morphology but also the specific pathology present[36]. While valve replacement may neutralize the anatomical disadvantages of bicuspid valves, repair techniques must work within the constraints of the existing anatomy[38]. The substantial heterogeneity observed in procedural time outcomes (e.g., ACX time: I² = 93%; CPB time: I² = 94%) in this meta-analysis likely stems from multiple sources, including variations in surgical technique, patient anatomy, institutional experience, and the learning curve associated with BAV-specific repairs. Differences in surgical approaches, such as the choice of annuloplasty techniques or valve-sparing procedures, may contribute significantly to variability in procedural times, as BAV repairs often require tailored strategies to address complex anatomical variations like bicuspid valve morphology or asymmetric commissural orientation. Patient-specific factors, such as aortic root dimensions or leaflet calcification, may further complicate repairs and prolong operative times. Additionally, institutional experience and surgeon familiarity with BAV-specific techniques likely play a critical role, as centers with higher case volumes may achieve more consistent outcomes due to refined protocols and expertise. The learning curve for BAV repairs, which demand precise intraoperative assessments like effective height measurement or commissural orientation analysis, may also contribute to heterogeneity, particularly in less experienced centers. To address these challenges, standardized surgical approaches and outcome reporting are essential. The AVIATOR Registry, supported by the Heart Valve Society, represents a pivotal step toward achieving this goal by collecting homogeneous data across multiple centers[39,40]. Future research should focus on identifying optimal surgical techniques for different anatomical variants, establishing evidence-based selection criteria, and promoting the universal adoption of objective assessment tools, such as effective height measurement and systematic commissural orientation analysis, to enhance repair quality, durability, and consistency across institutions.

Clinical implications

The findings support the use of VSARR as a safe option for both BAV and TAV patients in experienced centers, offering the benefit of avoiding prosthetic valve complications, such as anticoagulation-related risks and thromboembolism[3,4]. However, the higher reintervention risk in BAV patients necessitates meticulous patient selection and detailed preoperative counseling, particularly for younger patients who may prioritize anticoagulation avoidance but face a greater likelihood of future surgeries[30,32]. Surgeons should clearly communicate the trade-offs between VSARR and valve replacement, emphasizing the need for long-term monitoring, especially in BAV patients. Referral to high-volume centers with specialized aortic valve expertise is recommended for complex BAV cases to optimize outcomes. Enhanced surveillance protocols, including more frequent echocardiographic evaluations for BAV patients beyond five years, are critical to detect early valve deterioration and guide timely interventions.

Limitations

This meta-analysis has some limitations. First, the included studies were only observational, with no randomized controlled trials available, introducing potential confounding and selection bias that may affect the validity of the findings, despite the low-to-moderate risk of bias in most included studies. Second, substantial heterogeneity in procedural time outcomes reflects variability in surgical techniques, institutional practices, and reporting, which limits the precision of these findings. Third, incomplete reporting of patient characteristics, such as preoperative AR severity and comorbidities, as well as unmeasured anatomical variables like leaflet symmetry, commissural orientation, and aortic root dimensions, which can significantly influence clinical outcomes, precluded comprehensive pooled analyses of potential confounders. Fourth, the predominance of data from high-volume centers may reduce generalizability to less specialized settings. Finally, evidence of publication bias in procedural time outcomes suggests possible underrepresentation of smaller studies with non-significant results, which may affect the robustness of these findings.

Future directions

Future research should prioritize standardizing VSARR techniques to minimize heterogeneity and enhance outcomes, particularly for BAV patients. Prospective, multicenter studies with uniform outcome reporting, such as those supported by the AVIATOR Registry, are essential to elucidate long-term valve durability and refine patient selection criteria[39,40]. Exploring the utility of advanced imaging modalities, such as 4D flow MRI, could improve early detection of valve deterioration and inform personalized surveillance strategies[26]. Comparative studies of VSARR vs valve replacement in BAV patients across diverse populations and surgical settings are needed to develop evidence-based treatment algorithms. Additionally, investigating innovative repair techniques, such as advanced annuloplasty or tissue-engineered grafts, may improve the durability of BAV repairs[28,29].

CONCLUSION

This meta-analysis of 13 studies, encompassing 1419 BAV and 2349 TAV patients, establishes VSARR as a safe and viable option for both valve morphologies, with comparable short-term mortality and procedural complexity. However, the significantly higher long-term reintervention risk in BAV patients highlights the need for tailored surgical strategies, rigorous long-term surveillance, and referral to specialized centers. These findings emphasize the importance of individualized decision-making and underscore the necessity for standardized techniques and further research to optimize outcomes in BAV patients undergoing VSARR.