Early and late outcomes of surgical aortic valve replacement with or without aortic annular enlargement in patients with small aortic annulus: an exploratory comparison of Y-incision versus Manouguian techniques

Abstract

Background

Whether aortic annular enlargement (AAE) improves outcomes during surgical aortic valve replacement (SAVR) in patients with small aortic annulus (SAA) remains unclear. This study compared early and late outcomes of SAVR with or without AAE and evaluated the Y-incision versus Manouguian techniques.

Methods

We retrospectively analyzed 255 patients with SAA undergoing SAVR between January 2013 and March 2025. The cohort was classified into patients receiving isolated SAVR (AVR group, n=187) or SAVR with AAE (AVR + AAE group, n=68), with the latter further stratified by technique into the Manouguian (n=37) and Y-incision (n=31) subgroups. Propensity score matching (PSM) was performed. Survival was assessed using Kaplan-Meier analysis, and readmission and reoperation were evaluated using Fine-Gray competing risk models.

Results

After matching, all baseline demographics, comorbidities, and surgical indications were comparable between the groups. The AVR + AAE group had longer cardiopulmonary bypass time and cross-clamp time but received larger prostheses with higher effective orifice area index (EOAI) and lower rates of moderate-to-severe prosthesis-patient mismatch (PPM), demonstrating superior hemodynamics across all follow-ups. Despite longer intensive care unit stays and higher rates of prolonged ventilation, reintubation, and dialysis-requiring acute kidney injury, 30-day mortality and 5-year survival were comparable (95.6% vs. 97.4%, P=0.47). AAE independently reduced cardiac-related readmission, with a lower 5-year cumulative incidence (5.4% vs. 19.5%, P=0.028). Within the AAE cohort, Y-incision achieved greater annular enlargement, higher EOAI, and lower PPM incidence, with similar early outcomes and 2-year event-free survival (P=0.84).

Conclusions

AAE should be considered during SAVR for SAA patients to optimize valve hemodynamics and reduce readmissions. The Y-incision provides greater annular enlargement and hemodynamic benefits than the Manouguian technique, supporting its wider application in clinical practice.

Highlight box.

Key findings

• In patients with small aortic annulus (SAA), surgical aortic valve replacement (SAVR) with concomitant aortic annular enlargement (AAE) significantly improved postoperative valve hemodynamics and reduced the incidence of prosthesis-patient mismatch (PPM) without increasing early or late mortality.

• Concomitant AAE was independently associated with a lower risk of cardiac-related readmission during follow-up.

• In an exploratory sub-analysis, the Y-incision and Manouguian techniques showed comparable safety and short-term clinical outcomes, while the Y-incision achieved greater annular enlargement, superior hemodynamics, and a lower incidence of PPM.

What is known and what is new?

• PPM is common in patients with SAA and is associated with adverse hemodynamic and clinical outcomes. AAE can facilitate implantation of larger prostheses, but its impact on outcomes and the optimal technique remains controversial.

• This study provides a propensity score-matched comparison of SAVR with and without AAE specifically in patients with SAA, and offers a direct exploratory comparison between the Y-incision and Manouguian techniques.

What is the implication, and what should change now?

• For patients with SAA, concomitant AAE can be considered a safe strategy to improve postoperative hemodynamics and reduce cardiac-related readmission.

• The Y-incision technique may offer hemodynamic advantages over traditional enlargement methods without compromising safety, supporting its consideration in selected patients within a lifetime valve management framework.

Introduction

Background

Surgical aortic valve replacement (SAVR) is the standard treatment for severe aortic stenosis and/or regurgitation (1). In clinical practice, surgeons strive to implant the largest feasible prosthetic valve that the patient’s annulus can accommodate. However, the prosthesis’s struts and sewing ring reduce the effective orifice area (EOA) compared with the native valve (2). This discrepancy can lead to prosthesis-patient mismatch (PPM), a condition in which the EOA is insufficient for the patient’s body surface area (BSA), creating a functional stenosis with an elevated transvalvular pressure gradient (3,4). Notably, PPM is particularly prevalent among patients with small aortic annulus (SAA, defined as an intraoperative annular diameter ≤21 mm) and has been linked to adverse clinical outcomes (5,6).

For SAA patients undergoing SAVR, aortic annular enlargement (AAE) enables implantation of a larger prosthesis, which not only mitigates the risk of PPM but also facilitates potential future valve-in-valve transcatheter aortic valve replacement (TAVR). Traditional AAE techniques include the Nicks, Manouguian, and Konno procedures (7-10), while the Y-incision approach, introduced in 2020, represents a recent surgical innovation (11).

Rationale and knowledge gap

Nevertheless, it remains unclear whether AAE improves clinical outcomes in SAA patients, particularly concerning postoperative complications and survival. Existing comparative studies on SAVR with AAE specifically in SAA patients are limited, and many are confounded by the inclusion of patients with a normal annulus who did not require AAE. Furthermore, evidence regarding the safety and efficacy of the Y-incision technique remains particularly scarce, with a paucity of direct comparisons against traditional AAE procedures. This evidence gap poses a clinical dilemma for surgeons in selecting the surgical technique for SAA patients.

Objective

Therefore, we conducted a retrospective cohort study to compare the early and late clinical outcomes of SAA patients undergoing SAVR with or without AAE. Additionally, we performed a sub-analysis comparing the Y-incision and Manouguian AAE techniques, and identified independent risk factors for cardiac-related readmission following SAVR in this population. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0594/rc).

Methods

Patients

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of The First Affiliated Hospital with Nanjing Medical University (No. 2025-SR-1215) and individual consent for this retrospective analysis was waived. From January 2013 to March 2025, 255 patients with SAA underwent SAVR for aortic valve disease at The First Affiliated Hospital with Nanjing Medical University. Among them, 187 patients underwent isolated SAVR (AVR group, n=187), and 68 patients underwent SAVR with concomitant AAE (AVR + AAE group, n=68). Within the AVR + AAE group, Manouguian technique was applied in 37 patients (Manouguian group, n=37), while Y-incision was applied in 31 patients (Y-incision group, n=31). After excluding 13 patients with active infective endocarditis or missing data, optimal full matching was applied as the propensity score matching (PSM) algorithm to balance baseline characteristics, including 14 covariates: sex, age, BSA, hypertension, diabetes, coronary artery disease, chronic lung disease, end-stage renal disease, prior stroke, New York Heart Association (NYHA) functional class, prior cardiac surgery, bicuspid aortic valve, surgical indication, and ejection fraction (EF). To improve robustness, matching weights were truncated at the 99th percentile. The resulting effective sample sizes were 164 for the AVR group and 68 for the AVR + AAE group; the Manouguian and Y-incision groups had effective sample sizes of 34 and 31, respectively. A standardized mean difference (SMD) of <0.1 indicated adequate covariate balance (Figure 1).

Figure 1.

Love plots of standardized mean differences before and after propensity score matching. (A) Covariate balance for the AVR group versus AVR + AAE group. (B) Covariate balance for the Manouguian group versus Y-incision group. Standardized mean differences <0.1 indicate good covariate balance. AAE, aortic annular enlargement; AVR, aortic valve replacement; NYHA, New York Heart Association.

Interventions

Two AAE techniques were employed: the Manouguian and Y-incision procedures. The Y-incision technique was introduced at The First Affiliated Hospital with Nanjing Medical University in 2022, and the choice of performing AAE and the selection of technique were primarily based on surgeon preference. All operations were performed via median sternotomy under general anesthesia. In the Manouguian technique, the aortotomy was extended across the aortic annulus at the commissure between the left and non-coronary cusps into the aorto-mitral curtain to the level of the mitral annulus. When necessary, the incision could be further extended into the anterior mitral leaflet and the roof of the left atrium. The defect was reconstructed using a teardrop-shaped bovine pericardial patch (12). In the Y-incision technique, the aortotomy was extended through the same commissure and further divided in a Y-shaped fashion into the aorto-mitral curtain toward the left and right fibrous trigones. Reconstruction was completed using a rectangular bovine pericardial patch (13).

Study design

This retrospective cohort study compared early and late clinical outcomes of SAA patients undergoing SAVR with or without AAE. Early outcomes were defined as events occurring within 30 days after surgery or during the index hospitalization. Late outcomes were evaluated up to five years for the comparison between the AVR and AVR + AAE groups, and up to two years for the comparison between the Manouguian and Y-incision techniques due to the later adoption of the Y-incision technique.

The primary endpoints were all-cause mortality, cardiac-related readmission, and reoperation during follow-up. Secondary endpoints included postoperative, 3-month, and 12-month echocardiographic hemodynamics, as well as postoperative complications and operative outcomes.

Data collection

Clinical data covering the preoperative, intraoperative, and postoperative periods were obtained through a review of patient medical records. Transthoracic echocardiographic assessments were performed during the index hospitalization for both preoperative and postoperative evaluations, with follow-up echocardiograms conducted at three and 12 months postoperatively during scheduled outpatient visits. Late outcomes, including survival, reoperation, and cardiac-related readmission, were collected primarily from institutional medical records and supplemented by structured telephone interviews with patients or their families.

The collected echocardiographic data included aortic valve peak and mean pressure gradients, jet velocity, left ventricular diastolic diameter (LVDD), interventricular septum thickness (IVST), left ventricular posterior wall thickness (LVPWT), and EF. The left ventricular mass index (LVMI) was calculated using the following formula:

$$LVMI=\frac{0.8\times 1.04\times \left[{\left(LVDD+LVPWT+IVST\right)}^3-LVD{D}^3\right]+0.6}{BSA}$$ [1]

PPM was assessed using the effective orifice area index (EOAI), calculated as:

$$EOAI=\frac{EOA}{BSA}$$ [2]

PPM severity was classified according to the current standard definitions: moderate PPM as EOAI <0.85 cm²/m², and severe PPM as EOAI <0.65 cm²/m² (4). The EOA values were obtained from manufacturer-provided labeled data.

Statistical analysis

All analyses were conducted using R (version 4.5.1). Optimal full matching on the propensity score was performed to balance baseline characteristics between groups (Figure 1), generating weights and subclasses for each case. Weights were truncated at the 99th percentile to minimize the influence of extreme values. After weighting, continuous variables with normal distributions were compared using t-test and presented as mean ± standard deviation, whereas non-normally distributed variables were compared using the Mann-Whitney U test and presented as median and interquartile range. Ordinal variables were compared using the Mann-Whitney U test and presented as frequency and percentage for each category. Categorical variables were compared using the Chi-squared or Fisher’s exact test and presented as frequency and percentage. Survival was analyzed using Kaplan-Meier methods, with group differences assessed by the Wald test from weighted Cox proportional hazards regression. Reoperation and cardiac-related readmission were analyzed using the cumulative incidence function with death treated as a competing risk, and group differences were evaluated using the Wald test based on the Fine-Gray subdistribution hazards model. Multivariable Fine-Gray regression was further performed to identify independent risk factors for cardiac-related readmission, with results reported as subdistribution hazard ratios (SHRs) and 95% confidence intervals (CIs). In the AAE sub-analysis, a 2-year composite endpoint of all-cause mortality, reoperation, or cardiac-related readmission was evaluated using Kaplan-Meier curves, with group differences assessed by the Wald test. A two-sided P value <0.05 was considered statistically significant.

Results

Comparison between AVR and AVR + AAE groups

Demographic and preoperative data

Demographic and preoperative characteristics are summarized in Table 1. After PSM, the overall matched cohort had a median age of 61 years, and 24.8% of patients were male. All baseline covariates were well balanced between groups, with SMDs below 0.1 (Figure 1A). Preoperative echocardiographic hemodynamic data, including peak and mean pressure gradients and jet velocity, were comparable between the matched groups (Table 1).

Table 1. Demographic and preoperative data: AVR group vs. AVR + AAE group.

Variable	Before propensity score matching					After propensity score matching
	AVR (n=174)	AVR + AAE (n=68)	SMD	P value		AVR (n=164)	AVR + AAE (n=68)	SMD	P value
Sex (male)	65 (37.4)	17 (25.0)	0.124	0.09		40 (24.4)	17 (25.0)	0.003	0.86
Age, years	57 [48–64]	61 [55–67]	0.169	0.04		60 [50–67]	61 [55–67]	0.003	0.50
Body surface area, m2	1.65±0.17	1.63±0.16	0.112	0.43		1.64±0.20	1.63±0.16	0.014	0.96
Hypertension	53 (30.5)	24 (35.3)	0.048	0.57		69 (42.1)	24 (35.3)	0.067	0.62
Diabetes mellitus	8 (4.6)	7 (10.3)	0.057	0.18		6 (3.7)	7 (10.3)	0.065	0.050
Coronary artery disease	38 (21.8)	11 (16.2)	0.057	0.42		33 (20.1)	11 (16.2)	0.041	0.71
Prior myocardial infarction	0 (0.0)	2 (2.9)	0.029	0.08		0 (0.0)	2 (2.9)	0.029	0.09
Chronic lung disease	2 (1.1)	3 (4.4)	0.033	0.14		10 (6.1)	3 (4.4)	0.018	0.76
End-stage renal disease	1 (0.6)	3 (4.4)	0.038	0.07		1 (0.6)	3 (4.4)	0.036	0.08
Prior stroke	6 (3.4)	6 (8.8)	0.054	0.16		3 (1.8)	6 (8.8)	0.067	0.02
NYHA functional class				0.51					0.55
Class II	44 (25.3)	13 (19.1)	0.062			22 (13.4)	13 (19.1)	0.054
Class III	110 (63.2)	48 (70.6)	0.074			124 (75.6)	48 (70.6)	0.051
Class IV	20 (11.5)	7 (10.3)	0.012			17 (10.4)	7 (10.3)	0.003
Prior cardiac surgery	7 (4.0)	21 (30.9)	0.269	<0.001		42 (25.6)	21 (30.9)	0.051	0.92
Bicuspid aortic valve	17 (9.8)	17 (25.0)	0.152	0.004		51 (31.1)	17 (25.0)	0.064	0.56
Aortic valve stenosis	91 (52.3)	56 (82.4)	0.301	<0.001		126 (76.8)	56 (82.4)	0.055	0.55
Aortic valve insufficiency	144 (82.8)	41 (60.3)	0.225	<0.001		100 (61.0)	41 (60.3)	0.005	0.72
Case status (urgent)	4 (2.3)	2 (2.9)	0.006	0.67		1 (0.6)	2 (2.9)	0.026	0.21
Ejection fraction, %	63.3 [59.9–65.5]	63.6 [59.7–65.4]	0.198	0.94		63.7 [56.3–65.6]	63.6 [59.7–65.4]	0.036	0.81
LVMI, g/m2	122.13 [96.66–154.66]	122.04 [97.26–154.56]	–	0.80		123.25 [91.00–158.70]	122.04 [97.26–154.56]	–	0.75
Echocardiographic data
Peak gradient, mmHg	48.0 [30.0–70.0]	71.0 [40.0–86.0]	–	<0.001		50.0 [39.0–86.0]	71.0 [40.0–86.0]	–	0.24
Mean gradient, mmHg	25.0 [17.0–38.0]	41.0 [22.0–52.0]	–	<0.001		27.0 [22.0–48.0]	41.0 [22.0–52.0]	–	0.13
Jet velocity, m/s	3.40 [2.70–4.00]	4.20 [3.20–4.70]	–	<0.001		3.50 [3.10–4.60]	4.20 [3.20–4.70]	–	0.23

Values are presented as mean ± SD or median [interquartile range] for continuous variables and n (%) for categorical variables. AAE, aortic annular enlargement; AVR, aortic valve replacement; LVMI, left ventricular mass index; NYHA, New York Heart Association; SD, standard deviation; SMD, standardized mean difference.

Intraoperative data

The use of mechanical valves was more frequent in the AVR group than in the AVR + AAE group (42.1% vs. 13.2%; P<0.001). Cardiopulmonary bypass time (157 vs. 143 minutes; P<0.001) and cross-clamp time (128 vs. 120 minutes; P=0.001) were significantly longer in the AVR + AAE group. The median native annular diameter measured intraoperatively was 19 mm in the AVR group and 20 mm in the AVR + AAE group, with no significant difference (P=0.58). In contrast, the AVR + AAE group received significantly larger prosthetic valves (median size, 23 vs. 19; P<0.001), achieved a higher median EOAI (1.08 vs. 0.85 cm²/m²; P<0.001), and exhibited a lower incidence of moderate-to-severe PPM (20.6% vs. 49.4%; P=0.002). Apart from ascending aorta procedures, no significant differences were observed in the types of concomitant procedures between the two groups (Table 2).

Table 2. Intraoperative & postoperative data: AVR group vs. AVR + AAE group.

Variable	AVR (n=164)	AVR + AAE (n=68)	P value
Cardiopulmonary bypass time, min	143 [105–163]	157 [136–204]	<0.001
Cross-clamp time, min	120 [72–130]	128 [107–165]	0.001
Prosthesis type (mechanical)	69 (42.1)	9 (13.2)	<0.001
Native annulus size, mm	19 [19–21]	20 [19–20]	0.58
Implanted valve size	19 [19–21]	23 [23–23]	<0.001
Number of sizes upsized	0 [0–0]	2 [2–2]	<0.001
Indexed effective orifice area, cm2/m2	0.85 [0.75–0.88]	1.08 [0.88–1.29]	<0.001
Moderate/severe patient-prosthesis mismatch	81 (49.4)	14 (20.6)	0.002
Concomitant procedures
Mitral valve procedure	93 (56.7)	32 (47.1)	0.30
Tricuspid valve procedure	66 (40.2)	19 (27.9)	0.08
CABG	18 (11.0)	4 (5.9)	0.33
Ascending/arch procedure	0 (0.0)	10 (14.7)	<0.001
Annular enlargement technique
Manouguian	–	37 (54.4)
Y-incision	–	31 (45.6)
Complication
Reoperation for bleeding	1 (0.6)	0 (0.0)	>0.99
Prolonged ventilation (>24 h)	6 (3.7)	17 (25.0)	<0.001
Tracheal reintubation	3 (1.8)	5 (7.4)	0.042
New dialysis	7 (4.3)	11 (16.2)	0.03
New-onset atrial fibrillation	25 (15.2)	8 (11.8)	0.64
Postoperative stroke	0 (0.0)	1 (1.5)	0.28
Pulmonary infection	9 (5.5)	6 (8.8)	0.44
Pleural effusion	34 (20.7)	10 (14.7)	0.18
Pacemaker implantation	5 (3.0)	3 (4.4)	0.69
Length of stay
Hospital, days	12 [9–15]	12 [9–15]	0.81
Intensive care unit, days	1 [1–2]	3 [2–6]	<0.001
30-day mortality	0 (0.0)	1 (1.5)	0.28
Postoperative echocardiography	n=164	n=66
Peak gradient, mmHg	33.0 [30.0–38.0]	24.0 [18.0–32.0]	<0.001
Mean gradient, mmHg	18.0 [16.0–20.0]	10.0 [8.5–15.0]	<0.001
Jet velocity, m/s	2.90 [2.70–3.10]	2.40 [2.10–2.80]	<0.001
Echocardiography at 3 months follow-up	n=142	n=66
Peak gradient, mmHg	33.0 [30.0–38.0]	20.0 [13.0–26.0]	<0.001
Mean gradient, mmHg	18.0 [16.0–20.0]	10.0 [7.0–13.0]	<0.001
Jet velocity, m/s	2.90 [2.70–3.00]	2.20 [1.80–2.50]	<0.001
Ejection fraction, %	64.0 [61.4–65.6]	63.0 [59.0–64.4]	0.066
LVMI, g/m2	107.38 [88.80–134.34]	98.29 [84.50–117.50]	0.28
Echocardiography at 12 months follow-up	n=164	n=50
Peak gradient, mmHg	36.0 [27.0–47.0]	20.0 [15.0–27.0]	<0.001
Mean gradient, mmHg	19.0 [13.0–27.0]	11.0 [8.0–14.0]	<0.001
Jet velocity, m/s	3.00 [2.60–3.50]	2.20 [2.00–2.60]	<0.001
Ejection fraction, %	64.0 [61.9–65.9]	63.0 [59.0–64.4]	0.074
LVMI, g/m2	97.86 [88.64–121.33]	94.97 [85.16–108.83]	0.60

Values are presented as median [interquartile range] or n (%). AAE, aortic annular enlargement; AVR, aortic valve replacement; CABG, coronary artery bypass grafting; LVMI, left ventricular mass index.

Early outcomes

There were no significant differences between the two groups in major perioperative complications, including reoperation for bleeding, new-onset atrial fibrillation, postoperative stroke, pulmonary infection, pleural effusion, or pacemaker implantation. Patients in the AVR + AAE group, however, had longer intensive care unit (ICU) stays (3 vs. 1 day; P<0.001) and higher rates of prolonged ventilation >24 hours (25.0% vs. 3.7%; P<0.001), tracheal reintubation (7.4% vs. 1.8%; P=0.042), and dialysis-requiring acute kidney injury (16.2% vs. 4.3%; P=0.03). 30-day mortality was comparable between groups (1.5% vs. 0.0%; P=0.28) (Table 2).

Echocardiographic outcomes

Echocardiographic assessments during the index hospitalization and at three and 12 months showed lower peak and mean pressure gradients and jet velocity in the AVR + AAE group compared with the AVR group (all P<0.001), while EF and LVMI did not differ significantly (Table 2).

Late outcomes

The median follow-up duration for late outcomes was 60 months. Kaplan-Meier analysis demonstrated comparable 5-year survival between the AVR + AAE and AVR groups [95.6% (95% CI: 89.5–100%) vs. 97.4% (95% CI: 93.8–100%); P=0.47] (Figure 2).

Figure 2.

Kaplan-Meier curves for 5-year survival between the AVR and AVR + AAE groups. Shaded areas represent the 95% CI. 5-year survival rates were 97.4% (95% CI: 93.8–100%) for the AVR group and 95.6% (95% CI: 89.5–100%) for the AVR + AAE group (P=0.47). AAE, aortic annular enlargement; AVR, aortic valve replacement; CI, confidence interval.

The 5-year cumulative incidence of cardiac-related readmission was significantly lower in the AVR + AAE group compared with the AVR group [5.4% (95% CI: 0.0–11.4%) vs. 19.5% (95% CI: 13.3–25.7%); P=0.03] (Figure 3). In Fine-Gray competing risk analysis, AVR + AAE was associated with a significantly lower risk of cardiac-related readmission in the unadjusted model (SHR, 0.32; 95% CI: 0.12–0.89; P=0.03), and this association remained significant after multivariable adjustment (adjusted SHR, 0.18, 95% CI: 0.039–0.81; P=0.03) (Table 3). Older age, preoperative end-stage renal disease, and prior stroke were independently associated with an increased risk of cardiac-related readmission (Figure 4).

Figure 3.

Cumulative incidence of 5-year cardiac-related readmission between the AVR and AVR + AAE groups. Shaded areas represent the 95% CI. 5-year cardiac-related readmission rates were 19.5% (95% CI: 13.3–25.7%) for the AVR group and 5.4% (95% CI: 0.0–11.4%) for the AVR + AAE group (P=0.03). AAE, aortic annular enlargement; AVR, aortic valve replacement; CI, confidence interval.

Table 3. Fine-Gray regression analysis of the association between AVR + AAE and 5-year cardiac-related readmission in adjusted models.

Cardiac-related readmission	Model 1				Model 2				Model 3
	SHR	95% CI	P value		SHR	95% CI	P value		SHR	95% CI	P value
AVR + AAE	0.32	0.12–0.89	0.03		0.30	0.10–0.87	0.03		0.18	0.039–0.81	0.03

Model l was unadjusted. Model 2 was adjusted for sex, age. Model 3 was adjusted for sex, age, body surface area, hypertension, diabetes, end-stage renal disease, coronary artery disease, stroke. AAE, aortic annular enlargement; AVR, aortic valve replacement; CI, confidence interval; SHR, subdistribution hazard ratio.

Figure 4.

Multivariate Fine-Gray competing risks regression for factors of 5-year cardiac-related readmission. Vertical line at SHR =1.0 indicates no effect. AAE, aortic annular enlargement; AVR, aortic valve replacement; CI, confidence interval; SHR, subdistribution hazard ratio.

During follow-up, no patients in the AVR + AAE group underwent reoperation for valve degeneration, while three patients in the AVR group underwent reoperation. The cumulative incidence of reoperation differed between groups (P<0.001).

Comparison between Manouguian and Y-incision groups

Demographic and preoperative data

Demographic and preoperative characteristics of the AAE cohort are summarized in Table 4. After PSM, all covariates were well balanced, with SMDs below 0.1 (Figure 1B). Preoperative echocardiographic hemodynamics were comparable between groups (Table 4).

Table 4. Demographic and preoperative data: Manouguian group vs. Y-incision group.

Variable	Before propensity score matching					After propensity score matching
	Manouguian (n=37)	Y-incision (n=31)	SMD	P value		Manouguian (n=34)	Y-incision (n=31)	SMD	P value
Sex (male)	9 (24.3)	8 (25.8)	0.015	>0.99		7 (20.6)	8 (25.8)	0.046	0.63
Age, years	61 (47–65)	61 (55–67)	0.346	0.39		64 (44–70)	61 (55–67)	0.074	0.71
Body surface area, m2	1.64±0.18	1.63±0.14	0.074	0.76		1.63±0.13	1.63±0.14	0.049	0.82
Hypertension	13 (35.1)	11 (35.5)	0.003	>0.99		10 (29.4)	11 (35.5)	0.066	0.53
Diabetes mellitus	2 (5.4)	5 (16.1)	0.107	0.23		4 (11.8)	5 (16.1)	0.055	0.73
Coronary artery disease	5 (13.5)	6 (19.4)	0.058	0.75		6 (17.6)	6 (19.4)	0.003	0.87
Prior myocardial infarction	0 (0.0)	2 (6.5)	0.065	0.20		0 (0.0)	2 (6.5)	0.065	0.22
Chronic lung disease	1 (2.7)	2 (6.5)	0.037	0.59		0 (0.0)	2 (6.5)	0.053	0.23
End-stage renal disease	2 (5.4)	1 (3.2)	0.022	>0.99		1 (2.9)	1 (3.2)	0.015	>0.99
Prior stroke	4 (10.8)	2 (6.5)	0.044	0.68		3 (8.8)	2 (6.5)	0.028	>0.99
NYHA functional class				0.31					0.67
Class II	6 (16.2)	7 (22.6)	0.064			6 (17.6)	7 (22.6)	0.049
Class III	26 (70.3)	22 (71.0)	0.007			26 (76.5)	22 (71.0)	0.054
Class IV	5 (13.5)	2 (6.5)	0.071			2 (5.9)	2 (6.5)	0.005
Prior cardiac surgery	9 (24.3)	12 (38.7)	0.144	0.31		15 (44.1)	12 (38.7)	0.052	0.53
Bicuspid aortic valve	10 (27.0)	7 (22.6)	0.044	0.89		9 (26.5)	7 (22.6)	0.036	0.93
Aortic valve stenosis	29 (78.4)	27 (87.1)	0.087	0.52		31 (91.2)	27 (87.1)	0.059	0.20
Aortic valve insufficiency	25 (67.6)	16 (51.6)	0.160	0.28		19 (55.9)	16 (51.6)	0.049	0.60
Case status (urgent)	2 (5.4)	0 (0.0)	0.054	0.50		0 (0.0)	0 (0.0)	0.010	>0.99
Ejection fraction, %	63.3 (56.2–65.8)	63.0 (60.2–64.8)	0.060	0.99		61.4 (59.9–65.2)	63.0 (60.2–64.8)	0.098	0.52
LVMI, g/m2	129.12 (103.30–161.41)	114.48 (93.59–146.60)	–	0.22		129.12 (95.69–166.62)	114.48 (93.59–146.60)	–	0.55
Echocardiographic data
Peak gradient, mmHg	65.0 (36.0–94.0)	71.0 (63.0–83.0)	–	0.72		78.0 (44.0–134.0)	71.0 (63.0–83.0)	–	0.54
Mean gradient, mmHg	38.0 (22.0–56.0)	41.0 (32.0–51.0)	–	0.92		49.0 (25.0–63.0)	41.0 (32.0–51.0)	–	0.36
Jet velocity, m/s	4.00 (3.00–4.90)	4.20 (4.00–4.60)	–	0.65		4.40 (3.30–5.70)	4.20 (4.00–4.60)	–	0.63

Values are presented as mean ± SD or median (interquartile range) for continuous variables and n (%) for categorical variables. LVMI, left ventricular mass index; NYHA, New York Heart Association; SD, standard deviation; SMD, standardized mean difference.

Intraoperative data

The Y-incision group had a shorter cardiopulmonary bypass time than the Manouguian group (150 vs. 174 minutes; P=0.03), whereas cross-clamp time did not differ significantly (123 vs. 133 minutes; P=0.16). The median native annular diameter was 20 mm in both groups (P=0.16). However, the extent of annular enlargement differed, with the Y-incision group allowing implantation of prosthetic valves with a median increase of two sizes (up to three sizes in selected cases), compared with a one-size increase in the Manouguian group (P<0.001). As a result, the Y-incision group received larger implanted prostheses (maximum size, 27 vs. 25; P=0.046), demonstrated a higher median EOAI (1.25 vs. 0.94 cm²/m²; P=0.001), and exhibited a lower incidence of moderate-to-severe PPM (3.2% vs. 29.4%; P=0.009). Apart from coronary artery bypass grafting, there were no significant differences in the types of concomitant procedures. Other intraoperative variables, including prosthetic valve type and red blood cell transfusion, were comparable between the two groups (Table 5).

Table 5. Intraoperative & postoperative data: Manouguian group vs. Y-incision group.

Variable	Manouguian (n=34)	Y-incision (n=31)	P value
Cardiopulmonary bypass time, min	174 [150–200]	150 [132–170]	0.03
Cross-clamp time, min	133 [118–157]	123 [103–129]	0.16
Prosthesis type (mechanical)	6 (17.6)	3 (9.7)	0.49
Native annulus size, mm	20 [19–21]	20 [19–20]	0.16
Implanted valve size	23 [21–23]	23 [23–23]	0.046
Number of sizes upsized	1 [1–2]	2 [2–2]	<0.001
Indexed effective orifice area, cm2/m2	0.94 [0.82–1.08]	1.25 [1.15–1.38]	0.001
Moderate/severe patient-prosthesis mismatch	10 (29.4)	1 (3.2)	0.009
Concomitant procedures
Mitral valve procedure	16 (47.1)	8 (25.8)	0.27
Tricuspid valve procedure	8 (23.5)	8 (25.8)	0.79
CABG	0 (0.0)	3 (9.7)	0.09
Ascending/arch procedure	3 (8.8)	5 (16.1)	0.45
Red blood cell transfusion
Transfusion required	23 (67.6)	17 (54.8)	0.25
Units transfused, U	2.0 [0.0–4.0]	1.5 [0.0–4.0]	0.39
Complication
Reoperation for bleeding	0 (0.0)	0 (0.0)	–
Prolonged ventilation (>24 h)	6 (17.6)	4 (12.9)	0.75
Tracheal reintubation	3 (8.8)	2 (6.5)	>0.99
New dialysis	8 (23.5)	1 (3.2)	0.03
New-onset atrial fibrillation	4 (11.8)	3 (9.7)	>0.99
Postoperative stroke	2 (5.9)	0 (0.0)	0.50
Pulmonary infection	4 (11.8)	1 (3.2)	0.37
Pleural effusion	2 (5.9)	3 (9.7)	0.65
Pacemaker implantation	1 (2.9)	1 (3.2)	>0.99
Length of stay
Hospital, days	12 [9–23]	10 [8–12]	0.20
Intensive care unit, days	4 [2–6]	3 [2–5]	0.37
30-day mortality	0 (0.0)	0 (0.0)	>0.99
Postoperative echocardiography	n=32	n=31
Peak gradient, mmHg	26.0 [25.0–32.0]	19.0 [14.0–26.0]	0.005
Mean gradient, mmHg	15.0 [12.0–17.0]	10.0 [8.0–12.0]	0.003
Jet velocity, m/s	2.56 [2.50–2.90]	2.20 [1.90–2.50]	0.003
Echocardiography at 3 months follow-up	n=32	n=31
Peak gradient, mmHg	30.0 [23.0–37.0]	14.0 [12.0–19.0]	<0.001
Mean gradient, mmHg	16.0 [11.0–19.0]	7.00 [6.0–10.0]	<0.001
Jet velocity, m/s	2.80 [2.40–3.00]	1.90 [1.70–2.20]	<0.001
Ejection fraction, %	64.4 [60.1–64.8]	62.7 [59.0–64.0]	0.43
LVMI, g/m2	94.86 [90.20–103.78]	89.18 [83.88–113.18]	0.40
Echocardiography at 12 months follow-up	n=20	n=21
Peak gradient, mmHg	27.0 [20.0–38.0]	16.0 [14.0–20.0]	0.006
Mean gradient, mmHg	13.0 [11.0–24.0]	9.0 [7.0–10.0]	0.01
Jet velocity, m/s	2.60 [2.20–3.10]	2.00 [1.80–2.20]	0.005
Ejection fraction, %	63.0 [60.0–64.4]	63.3 [60.2–64.4]	0.98
LVMI, g/m2	98.81 [91.93–108.93]	90.27 [80.71–106.85]	0.25

Values are presented as median [interquartile range] or n (%). CABG, coronary artery bypass grafting; LVMI, left ventricular mass index.

Early outcomes

Most perioperative complications, as well as postoperative ICU and hospital length of stay, were comparable between groups. However, the incidence of postoperative dialysis-requiring acute kidney injury was significantly lower in the Y-incision group than in the Manouguian group (3.2% vs. 23.5%; P=0.03). No deaths occurred within 30 days in either group (Table 5).

Echocardiographic outcomes

Echocardiography performed during the index hospitalization, at three and 12 months showed lower transvalvular gradients and jet velocities in the Y-incision group, with no significant differences in EF or LVMI (Table 5).

Late outcomes

The median follow-up for late outcomes was 24 months. Freedom from the 2-year composite endpoint of all-cause mortality, reoperation, or cardiac-related readmission was comparable between the Manouguian and Y-incision groups [94.6% (95% CI: 87.1–100%) vs. 96.8% (95% CI: 90.8–100%); P=0.84] (Figure 5).

Figure 5.

Kaplan-Meier curves for 2-year freedom from the composite endpoint (all-cause mortality, reoperation, or cardiac-related readmission) between the Manouguian and Y-incision groups. Shaded areas represent the 95% CI. Two-year event-free rates were 94.6% (95% CI: 87.1–100%) for the Manouguian group and 96.8% (95% CI: 90.8–100%) for the Y-incision group (P=0.84). CI, confidence interval.

Similarly, the 2-year cumulative incidence of cardiac-related readmission did not differ significantly (P=0.28). Excluding perioperative deaths, no patients in either group experienced reoperation for valve degeneration or death during follow-up.

Discussion

In this study, we found that although SAVR with concomitant AAE was associated with higher rates of certain postoperative complications, there were no significant differences in 30-day mortality or 5-year survival compared with isolated SAVR. Importantly, AAE significantly improved postoperative aortic valve hemodynamics, reduced the incidence of PPM, and emerged as a protective factor against 5-year cardiac-related readmission. In the sub-analysis of AAE techniques, the Y-incision and Manouguian procedures demonstrated comparable early outcomes and 2-year composite endpoint event-free rates. However, the Y-incision technique achieved a greater degree of annular enlargement, resulting in superior hemodynamic performance and a lower incidence of PPM.

Our findings reaffirm the detrimental impact of PPM on clinical outcomes. In the AVR group, the higher incidence of moderate-to-severe PPM was accompanied by increased postoperative transvalvular gradients and jet velocities, reflecting a state of functional aortic stenosis. Previous studies have consistently shown that severe PPM is associated with persistent aortic stenosis-related symptoms, impaired left ventricular remodeling, arrhythmias, higher reintervention rates, and increased short-term and long-term mortality (3,4,14-18). Importantly, the clinical consequences of PPM are further amplified in the contemporary era of transcatheter valve therapies, as implantation of smaller initial surgical prostheses can result in severe residual PPM following valve-in-valve TAVR and compromise subsequent hemodynamic performance (19). Given the higher risk of PPM in patients with SAA, preventive strategies such as concomitant AAE are crucial, both for optimizing immediate hemodynamics and for preserving anatomical feasibility for future valve-in-valve TAVR.

Despite these potential advantages, whether concomitant AAE increases postoperative morbidity or mortality remains controversial, with prior studies reporting inconsistent results. Such discrepancies may largely be attributable to methodological limitations, including heterogeneous study populations not restricted to patients with SAA requiring AAE, as well as inadequate adjustment for baseline differences between treatment groups (20-27).

In our propensity-matched cohort, mechanical valve implantation was more frequent in the AVR group, likely reflecting an alternative surgical strategy adopted by some surgeons to mitigate PPM, as mechanical valves typically provide a larger EOA than bioprostheses of the same size (4,28). Nevertheless, the AVR + AAE group demonstrated superior hemodynamics, a higher EOAI, and a lower incidence of moderate-to-severe PPM, aligning with previous reports (29,30). Our data show that the AVR + AAE group had a longer ICU stay and a higher incidence of prolonged ventilation (>24 hours), tracheal reintubation, and dialysis-requiring acute kidney injury compared with the AVR group, whereas no significant differences were observed between the Y-incision and Manouguian techniques, suggesting that the increased early morbidity is attributable to the overall complexity of AAE rather than the specific technique. These findings are consistent with some prior studies and likely related to longer cardiopulmonary bypass time and more extensive manipulation of the aortic root (21,24,25). In line with this, cardiopulmonary bypass time and cross-clamp time were significantly longer in the AVR + AAE group than in the AVR group. Importantly, these complications did not translate into increased perioperative mortality, supporting the safety of the AAE procedure, and may represent a manageable trade-off for improved hemodynamics and a lower incidence of PPM. Although no significant 5-year survival benefit was observed, the protective association between AAE and reduced cardiac-related readmission underscores its potential long-term clinical benefit, in line with previous evidence (30).

To our knowledge, this is the first study to directly compare clinical outcomes between the Y-incision and Manouguian techniques. Prior studies have suggested that conventional techniques (Nicks or Manouguian) typically allow annular enlargement by one to two prosthesis sizes, whereas the Y-incision can achieve enlargement of two or more, even up to five sizes (2,11,31). At our center, the Y-incision technique was initially applied in a conservative manner, resulting in a median enlargement of two prosthesis sizes and a median prosthesis size of 23. With increasing experience, enlargement by three sizes and implantation of a 27-mm bioprosthesis were achieved. Despite this progression, a discrepancy remains compared with the results reported by Yang et al., which may partly reflect racial anatomical differences related to smaller BSA and annular dimensions in Asian patients (32). Notably, compared with the Manouguian group, the Y-incision group demonstrated superior enlargement, a larger EOAI, more significant hemodynamic advantages, and a lower incidence of moderate-to-severe PPM, without compromising safety or increasing mortality. The fact that the Y-incision provides more potent enlargement without an incremental time penalty compared to conventional techniques further supports its role as an optimized approach within the routine framework of aortic root enlargement.

Although our data did not demonstrate a direct survival benefit associated with improved hemodynamics in the AVR + AAE or Y-incision groups, in the current era of lifetime management of aortic valve disease, implantation of a larger prosthesis confers important downstream advantages, including lower postoperative gradients, better hemodynamics, and reduced PPM risk, which may not only slow the degeneration and extend the lifespan of bioprosthetic valves but also provide a more favorable anatomical foundation for potential future valve-in-valve TAVR.

This study has several limitations. As a single-center retrospective cohort study, selection and information bias cannot be fully excluded. Although PSM was applied to balance baseline characteristics, residual confounding from unmeasured factors–such as surgeon experience, procedural preference, and temporal bias related to the adoption of the Y-incision technique–may persist. The modest sample size, particularly for subgroup comparisons, may have limited statistical power, and follow-up for the Y-incision cohort remains relatively short. In addition, PPM was determined using manufacturer-reported in vitro EOA data rather than patient-specific in vivo measurements. Future large-scale, multi-center, prospective randomized controlled trials are warranted to further clarify the optimal role and extent of annular enlargement in patients with SAA.

Conclusions

Concomitant AAE during SAVR is a safe and effective strategy for patients with SAA, improving postoperative valve hemodynamics and reducing the risk of cardiac-related readmission. Compared to the Manouguian technique, the Y-incision technique provides superior enlargement and hemodynamic benefits with comparable safety.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of The First Affiliated Hospital with Nanjing Medical University (No. 2025-SR-1215) and individual consent for this retrospective analysis was waived.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0594/dss

jtd-18-06-641-dss.pdf ^{(68.9KB, pdf)}

DOI: 10.21037/jtd-2026-0594