Abstract
Objective
To compare perioperative and long-term outcomes between bicuspid (BAV) and tricuspid (TAV) aortic valve patients undergoing elective ascending aortic aneurysm surgery using propensity score matching (PSM).
Methods
We retrospectively analyzed 1340 consecutive adults who underwent elective ascending aortic surgery for aneurysm at a single high-volume center between January 2014 and December 2025. After exclusions, 816 patients (373 BAV, 443 TAV) were eligible. PSM (1:1 nearest-neighbor, caliper width 0.2 SD on the logit scale; 14 baseline covariates, including ascending aortic diameter) produced 285 matched pairs (all with standardized mean difference <0.10). Primary endpoints were 30-day and 10-year all-cause mortality; secondary endpoints included early postoperative complications and late aortic reintervention.
Results
The median follow-up was 5.4 years (interquartile range, 2.6-8.1; completeness, 96.7%). Thirty-day mortality was 0.7% (n = 2 of 285) in the BAV group versus 1.05% (n = 3 of 285) in the TAV group (odds ratio, 0.66; 95% confidence interval, [CI], 0.11-3.98; P = 1.00). Ten-year survival was 61.4% (95% CI, 55.8%-67.1%) in the BAV group versus 57.9% (95% CI, 52.2%-63.6%) in the TAV group (log-rank χ2 = 2.39; df = 1; P = .31). Early postoperative complications did not differ between the 2 groups (all P > .05). The cumulative incidence of late aortic reintervention at 10 years was 4.6% (95% CI, 2.1%-8.5%) in the BAV group versus 3.5% (95% CI, 1.4%-7.0%) in the TAV group (Gray test, P = .57). In multivariable Cox regression, BAV morphology was not an independent predictor of mortality (hazard ratio, 0.92; 95% CI, 0.58-1.45; P = .74).
Conclusions
In this propensity score–matched cohort, BAV morphology was not associated with worse perioperative or long-term outcomes after elective ascending aortic surgery compared to TAV. These findings support managing BAV-related aortopathy according to the same patient- and anatomy-driven principles used for TAV rather than based on valve morphology alone.
Key Words: bicuspid aortic valve, tricuspid aortic valve, ascending aortic aneurysm, propensity score matching, perioperative mortality, long-term survival
Graphical Abstract
Propensity score–matched survival analysis: BAV versus TAV in ascending aortic aneurysm surgery.
Central Message.
Bicuspid aortic valve morphology does not independently increase perioperative or 10-year mortality after elective ascending aortic aneurysm surgery.
Perspective.
In 285 propensity score–matched pairs undergoing elective ascending aortic aneurysm surgery, bicuspid and tricuspid valve morphology yielded equivalent 30-day mortality (0.7% vs 1.05%) and 10-year survival (61.4% vs 57.9%; log-rank P = .31), with comparable late reintervention rates. When patients are well selected and treated according to current guidelines, valve morphology alone should not drive operative strategy.
Bicuspid aortic valve (BAV) is the most common congenital cardiac anomaly, affecting 0.5% to 2% of the general population.1,2 It is frequently associated with aortopathy of the ascending aorta, with reported prevalence ranging from 20% to 84%.3,4 Population-based imaging studies suggest that aortic dilatation is detectable in approximately one-third of infants with BAV, indicating an early developmental component.5 Mutations in NOTCH1 and FBN1 have been implicated in both valve and aortic wall pathology.6,7
Although BAV patients tend to present for surgery at a younger age compared to those with tricuspid aortic valve (TAV) disease,2 contemporary cohorts increasingly demonstrate that BAV-associated aortopathy can be managed with outcomes comparable to those seen in TAV.8, 9, 10 Nevertheless, much of the available comparative evidence is limited by small sample sizes, lack of rigorous covariate balancing, and a primary focus on valve rather than aortic surgery.8,11
Against this background, the clinically relevant question is no longer whether BAV morphology should be considered inherently higher risk, but rather how outcomes of BAV and TAV ascending aortic surgery compare when measured carefully and adjusted for baseline risk. We therefore conducted this single-center, propensity score–matched analysis comparing BAV and TAV patients undergoing elective ascending aortic aneurysm surgery, with 10-year follow-up and explicit attention to ascending aortic diameter, surgical strategy, and late aortic reintervention.12
Methods
Study Design and Patient Population
This retrospective single-center cohort study analyzed all consecutive adults (age ≥18 years) who underwent elective ascending aortic surgery for aneurysm at the Department of Cardiac Surgery, Medical University of Silesia between January 2014 and December 2025. The study was approved by the Institutional Review Board of the Medical University of Silesia (PCN/CBN/0052/KB/105/23) with a waiver of individual consent for retrospective data collection and was conducted in accordance with the Declaration of Helsinki.
Aortic valve morphology was determined intraoperatively by direct surgical inspection and corroborated by preoperative transthoracic and transesophageal echocardiography and/or computed tomography angiography. Patients were excluded if they presented with emergency, urgent, or salvage indications; underwent concomitant mitral or tricuspid valve surgery (concomitant aortic valve replacement, repair, or root procedures were not an exclusion); had active infective endocarditis; had any prior cardiac surgery; or had missing baseline data required for propensity score calculation or insufficient follow-up for mortality assessment. Surgical indications followed contemporary guideline thresholds (generally ascending aortic diameter ≥55 mm [or ≥50 mm in selected BAV patients with additional risk features], family history of dissection, or rapid growth).13, 14, 15, 16
Propensity Score Matching
Of the 1340 screened patients, 524 were excluded based on prespecified criteria, yielding 816 eligible patients (373 BAV, 443 TAV). A logistic regression model with BAV morphology (0/1) as the dependent variable and 14 baseline covariates as predictors was used to estimate the propensity score: age, male sex, body mass index, left ventricular ejection fraction, New York Heart Association functional class, prior myocardial infarction, prior percutaneous coronary intervention, diabetes mellitus, hypertension, chronic kidney disease, chronic obstructive pulmonary disease, concomitant coronary artery bypass grafting, moderate-severe aortic regurgitation (grade ≥2), and maximum ascending aortic diameter on preoperative imaging.
Patients were matched 1:1 without replacement using the nearest-neighbor algorithm with a caliper width of 0.2 standard deviation (SD) on the logit-transformed propensity score (corresponding to approximately 0.04 on the original probability scale, given a logit SD of 0.83 in our cohort).17, 18, 19 Match order was randomized to minimize systematic bias. Balance was assessed using absolute standardized mean difference (SMD), with an SMD <0.10 considered indicative of acceptable balance and used as the primary balance criterion, with P values reported only as supporting information.20
Endpoints
The primary endpoints were 30-day all-cause mortality and long-term all-cause mortality. Thirty-day mortality was defined as death from any cause occurring within 30 days of the index surgical procedure, regardless of hospital status. Long-term mortality was assessed through comprehensive follow-up using the Polish National Health Fund, hospital database, and direct patient contact. Follow-up was censored at 10 years or the last known contact date.
Secondary endpoints included early postoperative complications: stroke (new neurologic deficit confirmed by imaging), acute kidney injury (RIFLE criteria: ≥50% increase in serum creatinine or urine output <0.5 mL/kg/hour for ≥6 hours),21 atrial fibrillation (new-onset during hospitalization), myocardial infarction (troponin elevation with electrocardiogram changes or imaging evidence), pneumonia (clinical and radiographic evidence), reexploration for bleeding (surgical reintervention for hemorrhage), and a composite of any early complication and late aortic reintervention (ie, any subsequent open or endovascular procedure on the thoracic aorta). Echocardiographic surveillance was performed routinely at discharge and at clinically indicated intervals; however, structured longitudinal echocardiographic data were not consistently available in the electronic record and are not reported in this analysis (see Limitations).
Follow-Up and Statistical Analysis
Long-term outcomes were ascertained through linkage with the Polish National Health Fund, the institutional clinical database, and direct telephone contact. Follow-up was censored at 10 years or the last known contact. The median follow-up (reverse Kaplan-Meier method) was 5.4 years (interquartile range [IQR], 2.6-8.1 years). Follow-up completeness, defined as the proportion of potential person-years actually observed, was 96.7%.
Continuous variables are presented as mean ± SD or median (IQR) and compared using the Student t test or Mann-Whitney U test, as appropriate. Categorical variables are presented as number (%) and compared using the χ2 or Fisher exact test. Long-term survival was estimated by the Kaplan-Meier method with 95% confidence intervals (CIs) derived from Greenwood's formula, and groups were compared using the Mantel-Cox (log-rank) test. Multivariable Cox proportional hazards regression was performed in the matched cohort as a sensitivity analysis, including BAV morphology and a parsimonious set of baseline covariates prespecified as biologically plausible mortality predictors (age, sex, left ventricular ejection fraction [LVEF], diabetes mellitus, chronic kidney disease, EuroSCORE II [European System for Cardiac Operative Risk Evaluation II]); the proportional hazards assumption was checked using Schoenfeld residuals, and discrimination was assessed using the Harrell C-index. Late aortic reintervention was analyzed using Fine-Gray competing risks regression with all-cause death as the competing event; cumulative incidence functions were compared using the Gray test. All tests were 2-sided with α = 0.05. Analyses were performed using R version 4.3.1 (packages MatchIt, survival, cmprsk) and SPSS Statistics 29.0 (IBM).
Results
Patient Selection and Propensity Score Matching
Of the 1340 consecutive screened patients, 524 were excluded (Figure 1), including 301 with an emergency/urgent/salvage presentation, 57 with concomitant mitral or tricuspid valve surgery, 78 with prior cardiac surgery, 21 with active infective endocarditis, 66 with missing baseline data or inadequate follow-up, and 1 age <18 years. The eligible cohort comprised 816 patients (373 BAV, 443 TAV). Propensity score matching (PSM) produced 285 1:1 matched pairs (570 patients), retaining 76.4% of BAV patients and 64.3% of TAV patients. All 14 covariates achieved an absolute SMD <0.10 in the matched cohort (Figure 2).
Figure 1.
CONSORT flow diagram of patient selection and propensity score matching.
Figure 2.
Covariate balance (absolute standardized mean difference) before and after propensity score matching for the 14 baseline covariates entered in the matching model. LVEF, Left ventricular ejection fraction; CABG, coronary artery bypass grafting; BMI, body mass index; COPD, chronic obstructive pulmonary disease; MI, myocardial infarction; NYHA, New York Heart Association; PCI, percutaneous coronary intervention.
Baseline Characteristics
Before PSM, compared to TAV patients, BAV patients were younger (mean age, 54.8 ± 13.1 years vs 60.9 ± 12.8 years; SMD = 0.68), more frequently male (76.7% vs 70.4%; SMD = 0.26), and had a more favorable comorbidity profile, including lower prevalences of diabetes (24.4% vs 29.8%), hypertension (65.7% vs 72.0%), and chronic kidney disease (7.5% vs 11.5%). The median preoperative ascending aortic diameter was 52 mm (IQR, 49-56 mm) in the BAV group and 53 mm (IQR, 50-57 mm) in the TAV group (SMD = 0.16). After PSM, all baseline characteristics were well balanced, including ascending aortic diameter (median, 52 mm [IQR, 49-56 mm] vs 53 mm [IQR, 50-57 mm]; SMD = 0.05). Detailed baseline characteristics before and after PSM are presented in Table 1.
Table 1.
Baseline characteristics before and after propensity score matching
| Characteristic | BAV (N = 373) |
TAV (N = 443) |
SMD∗ | BAV-PSM (N = 285) |
TAV-PSM (N = 285) |
SMD∗ |
|---|---|---|---|---|---|---|
| Demographics | ||||||
| Age, y, mean ± SD | 54.8 ± 13.1 | 60.9 ± 12.8 | 0.68 | 58.1 ± 12.5 | 58.5 ± 12.8 | 0.03 |
| Male sex, n (%) | 286 (76.7) | 312 (70.4) | 0.26 | 218 (76.5) | 209 (73.3) | 0.07 |
| Body mass index, kg/m2, mean ± SD | 26.2 ± 4.1 | 27.1 ± 4.3 | 0.18 | 26.3 ± 4.0 | 26.8 ± 4.2 | 0.04 |
| Cardiovascular history | ||||||
| Prior myocardial infarction, n (%) | 76 (20.4) | 108 (24.4) | 0.10 | 57 (20.0) | 58 (20.4) | 0.02 |
| Prior PCI, n (%) | 49 (13.1) | 71 (16.0) | 0.09 | 37 (13.0) | 38 (13.3) | 0.03 |
| Coronary artery disease, n (%) | ||||||
| None | 89 (23.9) | 76 (17.2) | 0.20 | 72 (25.3) | 71 (24.9) | 0.04 |
| Single vessel | 127 (34.1) | 142 (32.1) | 0.04 | 97 (34.0) | 97 (34.0) | 0.00 |
| Two vessel | 98 (26.3) | 127 (28.7) | 0.05 | 74 (26.0) | 76 (26.7) | 0.02 |
| Three vessel | 59 (15.8) | 98 (22.1) | 0.16 | 42 (14.7) | 41 (14.4) | 0.02 |
| Comorbidities, n (%) | ||||||
| Diabetes mellitus | 91 (24.4) | 132 (29.8) | 0.27 | 69 (24.2) | 71 (24.9) | 0.02 |
| Hypertension | 245 (65.7) | 319 (72.0) | 0.33 | 188 (65.9) | 191 (67.0) | 0.07 |
| Chronic kidney disease (any), n (%) | 28 (7.5) | 51 (11.5) | 0.24 | 22 (7.7) | 23 (8.1) | 0.04 |
| Stage 3a (GFR 45-59) | 18 (4.8) | 29 (6.5) | 0.07 | 14 (4.9) | 15 (5.3) | 0.02 |
| Stage 3b (GFR 30-44) | 8 (2.1) | 18 (4.1) | 0.10 | 6 (2.1) | 6 (2.1) | 0.00 |
| Stage 4 (GFR 15-29) | 2 (0.5) | 4 (0.9) | 0.04 | 2 (0.7) | 2 (0.7) | 0.00 |
| Chronic obstructive pulmonary disease | 41 (11.0) | 68 (15.3) | 0.13 | 31 (10.9) | 32 (11.2) | 0.03 |
| History of stroke | 12 (3.2) | 21 (4.7) | 0.07 | 9 (3.2) | 10 (3.5) | 0.02 |
| Peripheral arterial disease | 8 (2.1) | 19 (4.3) | 0.12 | 6 (2.1) | 7 (2.5) | 0.02 |
| Cardiac function | ||||||
| LVEF, %, mean ± SD | 52.4 ± 11.3 | 49.1 ± 13.2 | 0.21 | 51.8 ± 11.6 | 50.9 ± 12.8 | 0.06 |
| LVEF ≤35%, n (%) | 18 (4.8) | 29 (6.5) | 0.07 | 15 (5.3) | 16 (5.6) | 0.02 |
| NYHA class III-IV, n (%) | 54 (14.5) | 76 (17.2) | 0.09 | 37 (13.0) | 42 (14.7) | 0.04 |
| Aortic valve and aortic pathology | ||||||
| Aortic stenosis, n (%) | 156 (41.8) | 243 (54.9) | 0.27 | 119 (41.8) | 118 (41.4) | 0.01 |
| Aortic regurgitation grade ≥2, n (%) | 156 (41.8) | 198 (44.7) | 0.18 | 119 (41.8) | 120 (42.1) | 0.05 |
| Aortic valve area, cm2, mean ± SD | 1.2 ± 0.4 | 0.8 ± 0.3 | 1.13 | 1.2 ± 0.4 | 1.2 ± 0.4 | 0.00 |
| Mean transvalvular gradient, mm Hg, mean ± SD | 38.2 ± 21.6 | 51.4 ± 24.8 | 0.57 | 38.7 ± 21.9 | 38.1 ± 21.4 | 0.03 |
| Ascending aortic diameter, mm, median (IQR) | 52 (49-56) | 53 (50-57) | 0.16 | 52 (49-56) | 53 (50-57) | 0.05 |
| Operative planning | ||||||
| Concomitant CABG, n (%) | 142 (38.1) | 212 (47.9) | 0.20 | 108 (37.9) | 109 (38.2) | 0.01 |
| Risk assessment | ||||||
| EuroSCORE II, median (IQR) | 3.2 (2.1-5.1) | 3.8 (2.4-6.2) | 0.31 | 3.3 (2.2-5.0) | 3.4 (2.3-5.1) | 0.04 |
BAV, Bicuspid aortic valve; TAV, tricuspid aortic valve; SMD, standardized mean difference; PSM, propensity score matching; PCI, percutaneous coronary intervention; GFR, glomerular filtration rate (mL/min/1.73 m2); IQR, interquartile range; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; CABG, coronary artery bypass grafting; EuroSCORE II, European System for Cardiac Operative Risk Evaluation II.
SMD <0.10 indicates acceptable balance.
Operative Characteristics
In the matched cohort, the distribution of primary aortic procedures was similar in the BAV and TAV groups (Bentall, 13.0% vs 14.0%; valve-sparing root replacement, 11.6% vs 10.5%; supracoronary ascending aortic replacement, 27.4% vs 28.4%; hemiarch, 11.2% vs 9.8%; total arch, 5.3% vs 4.9%; all P > .05). Ascending aortic wrapping was more frequent in BAV patients (9.1% vs 2.5%; P < .01), reflecting tailored management of borderline ascending aortic diameters (50-53 mm) more often encountered in younger BAV patients. Concomitant aortic valve replacement (combined with ascending aortic replacement, for example, as part of a Bentall procedure or supracoronary replacement plus aortic valve replacement) was performed in 162 of 285 BAV patients (56.8%) and in 165 of 285 TAV patients (57.9%; P = .79); the remaining patients underwent valve-sparing root replacement or aortic valve repair. Cardiopulmonary bypass time (mean, 118.2 ± 38.4 minutes vs 122.1 ± 39.1 minutes; P = .29) and aortic cross-clamp time (mean, 68.4 ± 24.1 minutes vs 71.3 ± 23.8 minutes; P = .19) did not differ between the 2 groups. Detailed intraoperative data are reported in Table E1.
Primary Endpoint: 30-Day Mortality
Thirty-day all-cause mortality was 0.7% (n = 2 of 285) in the BAV group and 1.05% (n = 3 of 285) in the TAV group (odds ratio [OR], 0.66; 95% CI, 0.11-3.98; P = 1.00). The low and comparable rates indicate excellent operative safety in both groups.
Secondary Endpoints: Early Complications
No early postoperative complication differed significantly between groups. Forest plots of ORs and crude rates are shown in Figure 3, A. New-onset atrial fibrillation occurred in 16.5% of BAV versus 17.9% of TAV patients (OR, 0.91; 95% CI, 0.59-1.40; P = .74); acute kidney injury in 9.8% versus 10.9% (OR, 0.89; 95% CI, 0.52-1.53, P = .78); stroke in 2.8% versus 3.2% (OR, 0.89; 95% CI, 0.34-2.33; P = 1.00); and reexploration for bleeding in 2.5% versus 3.2% (P = .80). The composite any-complication rate was 27.4% versus 29.1% (P = .65). Postoperative complication rates are summarized in Figure 3, B.
Figure 3.
Postoperative complications in the matched cohort (285 pairs). A, Forest plot of odds ratios (bicuspid aortic valve [BAV] vs tricuspid aortic valve [TAV]) for each early complication. B, Crude event rates per group.
Long-Term Survival
With a median follow-up of 5.4 years (IQR, 2.6-8.1 years; completeness, 96.7%), Kaplan-Meier estimated survival was 99.3% for BAV versus 98.9% for TAV at 30 days, 96.5% (95% CI 94.4%-98.6%) versus 95.4% (95% CI, 93.0%-97.9%) at 1 year, 85.3% (95% CI, 81.1%-89.4%) versus 83.2% (95% CI, 78.8%-87.5%) at 5 years, and 61.4% (95% CI, 55.8%-67.1%) versus 57.9% (95% CI, 52.2%-63.6%) at 10 years (log-rank χ2 = 2.39; df = 1; P = .31). Survival curves and number at risk are shown in Figure 4. Timepoint estimates are presented in Table E2.
Figure 4.
Kaplan-Meier overall survival in the propensity score–matched cohort (n = 285 for both bicuspid aortic valve [BAV] and tricuspid aortic valve [TAV]) with a number at risk table. Shaded bands represent 95% confidence intervals (CIs) derived from Greenwood's formula; the type of CI is also indicated in the graph legend. Censored observations are marked with tick marks on each curve. Groups were compared using the Mantel-Cox (log-rank) test (χ2 = 2.39; df = 1; P = .31). Ten-year survival was 61.4% (95% CI, 55.8%-67.1%) in the BAV group versus 57.9% (95% CI, 52.2%-63.6%) in the TAV group.
Late Aortic Reintervention
During follow-up, 12 BAV patients (cumulative incidence, 4.6%; 95% CI, 2.1%-8.5% at 10 years) and 9 TAV patients (cumulative incidence, 3.5%; 95% CI, 1.4%-7.0% at 10 years) underwent late aortic reintervention (Gray test, P = .57). Indications included distal aortic progression (n = 11), proximal aortic redo (n = 5), and endovascular thoracic aortic repair for new pathology (n = 5). The reintervention rate did not differ significantly between the 2 groups.
Multivariable Cox Regression
In a parsimonious multivariable Cox model fitted in the matched cohort (sensitivity analysis), BAV morphology was not an independent predictor of mortality (hazard ratio [HR], 0.92; 95% CI, 0.58-1.45; P = .74). Independent predictors were age (HR, 1.03 per year; 95% CI, 1.02-1.05; P < .01), LVEF (HR, 0.98 per 1% increase; 95% CI, 0.97-0.99; P < .01), and EuroSCORE II (HR, 1.18 per point; 95% CI, 1.09-1.28; P < .01). The proportional hazards assumption was satisfied for all covariates (Schoenfeld test P > .05), and the Harrell C-index was 0.67. The complete model is shown in Table 2.
Table 2.
Multivariable Cox proportional hazards regression for long-term mortality (matched cohort, sensitivity analysis)
| Variable | HR | 95% CI | P value |
|---|---|---|---|
| BAV vs TAV | 0.92 | 0.58-1.45 | .74 |
| Age (per 1 y) | 1.03 | 1.02-1.05 | <.01 |
| Female sex | 1.20 | 0.82-1.77 | .34 |
| LVEF (per 1% increase) | 0.98 | 0.97-0.99 | <.01 |
| Diabetes mellitus | 1.36 | 0.96-1.94 | .08 |
| Chronic kidney disease | 1.46 | 0.99-2.17 | .06 |
| EuroSCORE II (per 1 point) | 1.18 | 1.09-1.28 | <.01 |
This parsimonious model included 7 prespecified covariates fitted in the matched cohort. The proportional hazards assumption was satisfied for all covariates (Schoenfeld residuals test P > .05); Harrell C-index = 0.67. HR, Hazard ratio; CI, confidence interval; BAV, bicuspid aortic valve; TAV, tricuspid aortic valve; LVEF, left ventricular ejection fraction; EuroSCORE II, European System for Cardiac Operative Risk Evaluation II.
Discussion
In this propensity score–matched analysis of 570 patients (285 pairs) undergoing elective ascending aortic surgery, the BAV and TAV groups had equivalent perioperative and long-term outcomes. Thirty-day mortality (0.7% vs 1.05%; P = 1.00) and 10-year survival (61.4% vs 57.9%; log-rank test, P = .31) were comparable, and the late aortic reintervention rate also was similar (4.6% vs 3.5% at 10 years; Gray test, P = .57). In multivariable analysis, morphology was not an independent predictor of mortality (HR, 0.92; P = .74). Traditional risk factors—age, LVEF, and EuroSCORE II—were the predominant determinants of long-term survival. These outcomes are consistent with recent reports on BAV patients undergoing surgery for type A aortic dissection.22
These findings also challenge historical concerns about increased surgical risk in BAV patients and provide robust evidence supporting current surgical guidelines.13, 14, 15, 16 The long-term survival profile in both groups aligns with contemporary reports of excellent late prognosis after elective aortic root and ascending aorta surgery.9,22, 23, 24 The small, nonsignificant survival advantage observed in BAV patients at 5 years and 10 years likely reflects their slightly more favorable risk profile and younger age rather than a protective effect of bicuspid morphology per se.9,25 Importantly, the absence of a morphology effect in multivariable analysis indicates that once age, ventricular function, and global operative risk are accounted for, cusp number does not meaningfully influence 10-year mortality. The nearly identical operative times, use of circulatory arrest, and rates of complex aortic procedures suggest that BAV anatomy can be managed with the same operative strategies and efficiency as TAV in experienced hands. The higher frequency of ascending aortic wrapping in BAV patients likely reflects tailored management of aortopathy rather than intrinsic procedural difficulty and did not translate into differences in early or late mortality.13,24
To our knowledge, this is one of the largest propensity score–matched analyses directly comparing BAV and TAV patients undergoing ascending aortic surgery with comprehensive long-term follow-up.8,9,22 Previous observational data have suggested increased complexity and potentially greater operative risk in BAV patients owing to anatomic variations, younger age at surgery, and different underlying pathophysiology1,13,26; however, recent multicenter studies have provided more nuanced perspectives. A large Swedish cohort study of 865 BAV patients undergoing aortic valve surgery found long-term relative survival equivalent to that of age- and sex-matched population controls, with no differences at 1 year, 5 years, 10 years, or 12 years, suggesting favorable outcomes when managed according to current guidelines.25 Similarly, a recent analysis of 747 BAV patients undergoing aortic root replacement for aneurysm reported in-hospital mortality of only 1.3% and 15-year survival of 79%, comparable to the general US population.23
The clinical importance of these findings is substantial, given the high prevalence of BAV (present in 1-2% of the general population) and the increased risk of aortic complications requiring surgical intervention.1,25 Historically, BAV has been considered a more aggressive disease process with earlier aortic dilation, higher dissection risk, and more complex surgical anatomy1,13,26; however, our results provide reassurance that with appropriate patient selection and modern surgical technique, BAV patients can achieve outcomes equivalent to TAV patients. A prospective study comparing bicuspid transcatheter aortic valve replacement and tricuspid surgical aortic valve replacement in low-risk patients found similar 1-year composite outcomes (4.2% for both groups), further supporting the equivalence of these treatment approaches across valve morphologies.27, 28, 29 These data are particularly valuable for surgical decision making, as they suggest that aortic valve morphology should not be an independent factor driving operative timing or technique selection and that decision making should focus on aortic diameter, hemodynamic severity, and patient-specific risk factors rather than morphology alone.
Several methodologic and clinical issues warrant emphasis. First, in the original version of the analysis, EuroSCORE II was included in the PSM model alongside several of its individual components, raising legitimate concerns about collinearity and overadjustment. The revised analysis presented here removes EuroSCORE II from the matching model and adds maximum ascending aortic diameter, arguably the most important preoperative anatomic determinant of operative indication and risk in this population. Reassuringly, the conclusions remained unchanged. Second, the surgical population is heterogeneous (Bentall, valve-sparing root, supracoronary, hemiarch and total arch replacement, and ascending aortic wrapping), and the greater use of wrapping in BAV patients (9.1% vs 2.5%) reflects tailored management of borderline aortic diameters in younger BAV patients rather than systematic differences in aortic management; the distribution of formal aortic replacement strategies did not differ between the 2 groups otherwise, and outcomes were unchanged in subgroup analyses excluding wrapping (data not shown). Third, beyond all-cause mortality, late aortic reintervention is a clinically meaningful outcome in BAV-related aortopathy. We observed comparable 10-year cumulative incidence (4.6% for BAV vs 3.5% for TAV), suggesting that with contemporary patient selection and surgical technique, the more aggressive aortopathy historically attributed to BAV does not translate into a substantially higher risk of reoperation in elective settings. However, these data should not be extrapolated to acute presentations such as type A aortic dissection, where outcomes in BAV remain an active area of investigation.10,22
Limitations
This study has several limitations. First, despite balanced PSM across 14 baseline variables—including ascending aortic diameter—its retrospective design is vulnerable to residual unmeasured confounding (eg, detailed aortic phenotyping, intraoperative hemodynamics, surgeon experience). Second, this is a single-center experience from a high-volume academic institution, which may limit the generalizability of our findings to lower-volume centers. Third, structured echocardiographic follow-up data on prosthetic valve function, left ventricular reverse remodeling, and evolution of distal aortic dimensions were not consistently available in our electronic record and could not be analyzed, limiting mechanistic insight into late valve- and aorta-related events. Fourth, we did not capture valve-related endpoints other than reintervention (eg, structural valve deterioration, endocarditis, prosthesis–patient mismatch) in a structured fashion, and only elective cases were included, and thus our findings should not be extrapolated to acute aortic syndromes.22,30
Conclusions
In this propensity score–matched, single-center cohort study, BAV and TAV morphologies were associated with comparable perioperative outcomes, comparable 10-year survival, and comparable rates of late aortic reintervention after elective ascending aortic surgery. Traditional risk factors—age, LVEF and EuroSCORE II—were the dominant determinants of long-term survival. These findings support a patient- and anatomy-driven approach to BAV-related aortopathy, applying the same indications and operative principles used for TAV, and should be confirmed in multicenter studies that incorporate structured echocardiographic and valve-related endpoints.
Conflict of Interest Statement
The authors reported no conflicts of interest.
The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.
Footnotes
This study was supported by the European Funds for Silesia 2021–2027 project “Supporting the transformation of the region by strengthening the potential of the Doctoral School of the Medical University of Silesia in Katowice”, under Agreement No. NWD/1028/2025.
The study was approved by the Institutional Review Board of the Medical University of Silesia (PCN/CBN/0052/KB/105/23, 15 June 2023). The requirement for individual patient consent was waived for retrospective data collection.
Appendix E1
Table E1.
Intraoperative characteristics in the propensity score–matched cohort
| Characteristic | BAV (N = 285) | TAV (N = 285) | P value |
|---|---|---|---|
| Primary aortic procedure, n (%) | |||
| Bentall procedure (composite graft) | 37 (13.0) | 40 (14.0) | .81 |
| Valve-sparing aortic root replacement (David/Yacoub) | 33 (11.6) | 30 (10.5) | .79 |
| Supracoronary ascending aortic replacement | 78 (27.4) | 81 (28.4) | .82 |
| Combined with concomitant AVR | 62 (21.8) | 67 (23.5) | .65 |
| Hemiarch replacement | 32 (11.2) | 28 (9.8) | .51 |
| Total arch replacement | 15 (5.3) | 14 (4.9) | .88 |
| Ascending aortic wrapping (external support) | 26 (9.1) | 7 (2.5) | <.01 |
| Other aortic procedure | 4 (1.4) | 5 (1.8) | .72 |
| Aortic valve management, n (%) | |||
| AVR + ascending aortic replacement (excl. Bentall) | 62 (21.8) | 67 (23.5) | .65 |
| AVR + Bentall | 37 (13.0) | 40 (14.0) | .81 |
| Valve-sparing root replacement (David/Yacoub) | 33 (11.6) | 30 (10.5) | .79 |
| Aortic valve repair | 86 (30.2) | 75 (26.3) | .35 |
| No valve intervention (preserved native valve) | 67 (23.5) | 73 (25.6) | .62 |
| Aortic procedures overall, n (%) | |||
| Ascending aorta replacement (any) | 285 (100.0) | 285 (100.0) | 1.00 |
| Aortic arch surgery | 3 (1.1) | 12 (4.2) | .04 |
| Concomitant procedures, n (%) | |||
| CABG | 108 (37.9) | 109 (38.2) | .92 |
| 1 graft | 31 (10.9) | 33 (11.6) | .76 |
| 2 grafts | 43 (15.1) | 41 (14.4) | .65 |
| 3 grafts | 22 (7.7) | 23 (8.1) | .90 |
| ≥4 grafts | 12 (4.2) | 12 (4.2) | 1.00 |
| AF ablation (Cox-Maze IV) | 1 (0.4) | 6 (2.1) | .13 |
| Operative times, mean ± SD | |||
| Aortic cross-clamp time, min | 68.4 ± 24.1 | 71.3 ± 23.8 | .19 |
| CPB time, min | 118.2 ± 38.4 | 122.1 ± 39.1 | .29 |
| Total operative time, min | 267.3 ± 71.2 | 271.4 ± 69.8 | .42 |
| Circulatory arrest | |||
| Use of circulatory arrest, n (%) | 67 (23.5) | 72 (25.3) | .57 |
| Circulatory arrest duration, min, mean ± SD | 32.1 ± 12.6 | 33.8 ± 13.2 | .23 |
| Lowest nasopharyngeal temperature, °C, mean ± SD | 28.4 ± 3.2 | 28.2 ± 3.4 | .69 |
| Prosthesis characteristics | |||
| Mechanical valve prosthesis, n (%) | 90 (45.0) | 75 (47.5) | .72 |
| Aortic conduit size, mm, mean ± SD | 31.3 ± 2.4 | 31.4 ± 3.0 | .41 |
| Aortic valve prosthesis size, mm, mean ± SD | 25.6 ± 2.9 | 26.3 ± 3.2 | .02 |
| Intraoperative management, n (%) | |||
| Reinitiation of CPB | 18 (6.3) | 21 (7.4) | .69 |
| Delayed sternal closure | 12 (4.2) | 15 (5.3) | .51 |
| Intraoperative pacemaker placement | 8 (2.8) | 9 (3.2) | .72 |
| Intraoperative IABP use | 14 (4.9) | 16 (5.6) | .69 |
| Vasoactive/inotropic support, n (%) | |||
| Any catecholamine infusion | 198 (69.5) | 206 (72.3) | .39 |
| Dopamine | 89 (31.2) | 98 (34.4) | .54 |
| Epinephrine | 124 (43.5) | 132 (46.3) | .46 |
| Norepinephrine | 67 (23.5) | 74 (26.0) | .31 |
| Milrinone | 43 (15.1) | 48 (16.8) | .53 |
| Dobutamine | 31 (10.9) | 36 (12.6) | .46 |
| Postoperative vasoactive support, n (%) | |||
| Postoperative IABP insertion | 28 (9.8) | 32 (11.2) | .88 |
| Length of stay, mean ± SD | |||
| ICU length of stay, d | 4.2 ± 3.1 | 4.4 ± 3.3 | .51 |
| Total length of stay, d | 8.3 ± 6.2 | 8.7 ± 6.4 | .39 |
| Transfusion requirements | |||
| RBC transfusion (any), n (%) | 218 (76.5) | 220 (77.2) | .80 |
| RBC units transfused, mean ± SD | 2.4 ± 1.3 | 2.6 ± 1.4 | .31 |
| Fresh frozen plasma transfusion, n (%) | 142 (49.8) | 148 (51.9) | .53 |
| Platelet transfusion, n (%) | 98 (34.4) | 104 (36.5) | .42 |
| Cryoprecipitate transfusion, n (%) | 24 (8.4) | 28 (9.8) | .69 |
BAV, Bicuspid aortic valve; TAV, tricuspid aortic valve; AVR, aortic valve replacement; CABG, coronary artery bypass grafting; AF, atrial fibrillation; SD, standard deviation; CPB, cardiopulmonary bypass; IABP, intra-aortic balloon pump; ICU, intensive care unit; RBC, red blood cells.
Table E2.
Long-term survival estimates with number at risk in the propensity score–matched cohort
| Time point | BAV survival, % (95% CI) | BAV n at risk | TAV survival, % (95% CI) | TAV n at risk | P value |
|---|---|---|---|---|---|
| 30 d | 99.3 (98.3-100.0) | 285 | 98.9 (97.8-100.0) | 285 | 1.00 |
| 1 y | 96.5 (94.4-98.6) | 268 | 95.4 (93.0-97.9) | 263 | .52 |
| 2 y | 93.5 (91.0-96.0) | 248 | 91.7 (88.9-94.5) | 242 | .41 |
| 3 y | 90.6 (87.7-93.5) | 224 | 88.5 (85.3-91.7) | 219 | .39 |
| 4 y | 87.8 (84.6-91.0) | 198 | 85.7 (81.9-89.5) | 192 | .45 |
| 5 y | 85.3 (81.1-89.4) | 169 | 83.2 (78.8-87.5) | 164 | .49 |
| 6 y | 80.5 (76.0-85.0) | 140 | 78.4 (73.7-83.1) | 134 | .52 |
| 7 y | 75.7 (70.8-80.6) | 109 | 73.6 (68.6-78.6) | 104 | .55 |
| 8 y | 71.0 (65.7-76.3) | 80 | 68.8 (63.4-74.2) | 76 | .57 |
| 9 y | 66.3 (60.7-71.9) | 53 | 63.6 (58.2-69.0) | 50 | .42 |
| 10 y | 61.4 (55.8-67.1) | 29 | 57.9 (52.2-63.6) | 27 | .31 |
Kaplan-Meier survival estimates with 95% confidence intervals (CIs) from Greenwood's formula. The number at risk decreased through follow-up because of deaths and administrative censoring (rolling enrolment between January 2014 and December 2025). Group comparisons are Mantel-Cox (log-rank) tests at each time point. BAV, Bicuspid aortic valve; TAV, tricuspid aortic valve.
References
- 1.Ward C. Clinical significance of the bicuspid aortic valve. Heart. 2000;83(1):81–85. doi: 10.1136/heart.83.1.81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Siu S.C., Silversides C.K. Bicuspid aortic valve disease. J Am Coll Cardiol. 2010;55(25):2789–2800. doi: 10.1016/j.jacc.2009.12.068. [DOI] [PubMed] [Google Scholar]
- 3.Verma S., Siu S.C. Aortic dilatation in patients with bicuspid aortic valve. N Engl J Med. 2014;370(20):1920–1929. doi: 10.1056/NEJMra1207059. [DOI] [PubMed] [Google Scholar]
- 4.Michelena H.I., Khanna A.D., Mahoney D., et al. Incidence of aortic complications in patients with bicuspid aortic valves. JAMA. 2011;306(10):1104–1112. doi: 10.1001/jama.2011.1286. [DOI] [PubMed] [Google Scholar]
- 5.Fernandes S.M., Sanders S.P., Khairy P., et al. Morphology of bicuspid aortic valve in children and adolescents. J Am Coll Cardiol. 2004;44(8):1648–1651. doi: 10.1016/j.jacc.2004.05.063. [DOI] [PubMed] [Google Scholar]
- 6.Garg V., Muth A.N., Ransom J.F., et al. Mutations in NOTCH1 cause aortic valve disease. Nature. 2005;437(7056):270–274. doi: 10.1038/nature03940. [DOI] [PubMed] [Google Scholar]
- 7.Prakash S.K., Bossé Y., Muehlschlegel J.D., et al. A roadmap to investigate the genetic basis of bicuspid aortic valve and its complications. J Am Coll Cardiol. 2014;64(8):832–839. doi: 10.1016/j.jacc.2014.04.073. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Holmgren A., Bull Enger T., Näslund U., et al. Long-term results after aortic valve replacement for bicuspid or tricuspid valve morphology in a Swedish population. Eur J Cardiothorac Surg. 2021;59(3):570–576. doi: 10.1093/ejcts/ezaa348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Girdauskas E., Disha K., Borger M.A., et al. Long-term prognosis of ascending aortic aneurysm after aortic valve replacement for bicuspid versus tricuspid aortic valve stenosis. J Thorac Cardiovasc Surg. 2014;147(1):276–282. doi: 10.1016/j.jtcvs.2012.11.004. [DOI] [PubMed] [Google Scholar]
- 10.Mennander A., Olsson C., Jeppsson A., et al. The significance of bicuspid aortic valve after surgery for acute type A aortic dissection. J Thorac Cardiovasc Surg. 2020;159(3):760–767.e3. doi: 10.1016/j.jtcvs.2019.03.012. [DOI] [PubMed] [Google Scholar]
- 11.Elbadawi A., Mahmoud A.A., Mahmoud K., et al. Temporal trends and outcomes of elective thoracic aortic repair and acute aortic syndromes in bicuspid aortic valves: insights from a national database. Cardiol Ther. 2021;10(2):531–545. doi: 10.1007/s40119-021-00237-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Benedetto U., Head S.J., Angelini G.D., et al. Statistical primer: propensity score matching and its alternatives. Eur J Cardiothorac Surg. 2018;53(6):1112–1117. doi: 10.1093/ejcts/ezy167. [DOI] [PubMed] [Google Scholar]
- 13.Borger M.A., Fedak P.W.M., Stephens E.H., et al. The American Association for Thoracic Surgery consensus guidelines on bicuspid aortic valve-related aortopathy. J Thorac Cardiovasc Surg. 2018;156(2):e41–e74. doi: 10.1016/j.jtcvs.2018.02.115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Isselbacher E.M., Preventza O., Hamilton Black J., et al. 2022 ACC/AHA guideline for the diagnosis and management of aortic disease. Circulation. 2022;146(24):e334–e482. doi: 10.1161/CIR.0000000000001106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Praz F., Borger M.A., Lanz J., et al. 2025 ESC/EACTS guidelines for the management of valvular heart disease. Eur Heart J. 2025;46(44):4635–4736. doi: 10.1093/eurheartj/ehaf194. [DOI] [PubMed] [Google Scholar]
- 16.Mazzolai L., Teixido-Tura G., Lanzi S., et al. 2024 ESC guidelines for the management of peripheral arterial and aortic diseases. Eur Heart J. 2024;45(36):3538–3700. doi: 10.1093/eurheartj/ehae179. [DOI] [PubMed] [Google Scholar]
- 17.Rosenbaum P.R., Rubin D.B. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70(1):41–55. [Google Scholar]
- 18.Nashef S.A.M., Roques F., Sharples L.D., et al. EuroSCORE II. Eur J Cardiothorac Surg. 2012;41(4):734–745. doi: 10.1093/ejcts/ezs043. [DOI] [PubMed] [Google Scholar]
- 19.Austin P.C. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46(3):399–424. doi: 10.1080/00273171.2011.568786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Austin P.C. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity score-matched samples. Stat Med. 2009;28(25):3083–3107. doi: 10.1002/sim.3697. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Bellomo R., Ronco C., Kellum J.A., et al. Acute renal failure: definition, outcome measures, animal models, fluid therapy and information technology needs. Crit Care. 2004;8(4):R204–R212. doi: 10.1186/cc2872. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kreibich M., Rylski B., Czerny M., et al. Type A aortic dissection in patients with bicuspid aortic valve aortopathy. Ann Thorac Surg. 2020;109(1):94–100. doi: 10.1016/j.athoracsur.2019.05.022. [DOI] [PubMed] [Google Scholar]
- 23.Rajesh K., Chung M., Levine D., et al. Long-term outcomes after aortic root replacement for bicuspid aortic valve-associated aneurysm. J Thorac Cardiovasc Surg. 2025;169(2):609–616.e4. doi: 10.1016/j.jtcvs.2024.03.003. [DOI] [PubMed] [Google Scholar]
- 24.Kim H.H., Lee S., Lee S.H., et al. The long-term fate of ascending aorta aneurysm after wrapping versus replacement. J Thorac Cardiovasc Surg. 2022;164(2):463–474.e4. doi: 10.1016/j.jtcvs.2020.12.110. [DOI] [PubMed] [Google Scholar]
- 25.Glaser N., Jackson V., Eriksson P., et al. Relative survival after aortic valve surgery in patients with bicuspid aortic valves. Heart. 2021;107(14):1167–1172. doi: 10.1136/heartjnl-2020-318733. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Della Corte A., Bancone C., Dialetto G., et al. The ascending aorta with bicuspid aortic valve: a phenotypic classification with potential prognostic significance. Eur J Cardiothorac Surg. 2014;46(2):240–247. doi: 10.1093/ejcts/ezt621. [DOI] [PubMed] [Google Scholar]
- 27.Mack M.J., Leon M.B., Thourani V.H., et al. Transcatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients. N Engl J Med. 2019;380(18):1695–1705. doi: 10.1056/NEJMoa1814052. [DOI] [PubMed] [Google Scholar]
- 28.Makkar R.R., Yoon S.H., Leon M.B., et al. Association between transcatheter aortic valve replacement for bicuspid vs tricuspid aortic stenosis and mortality or stroke. JAMA. 2019;321(22):2193–2202. doi: 10.1001/jama.2019.7108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kowalówka A.R., Kowalewski M., Wańha W., et al. Surgical and transcatheter aortic valve replacement for severe aortic stenosis in low-risk elective patients. J Thorac Cardiovasc Surg. 2024;167(5):1714–1723.e4. doi: 10.1016/j.jtcvs.2022.10.026. [DOI] [PubMed] [Google Scholar]
- 30.Davies R.R., Goldstein L.J., Coady M.A., et al. Yearly rupture or dissection rates for thoracic aortic aneurysms: simple prediction based on size. Ann Thorac Surg. 2002;73(1):17–28. doi: 10.1016/s0003-4975(01)03236-2. [DOI] [PubMed] [Google Scholar]