Continuous suture technique for aortic valve replacement in patients with small aortic annulus

Abstract

Background

The presence of a small aortic annulus (SAA) in patients undergoing aortic valve replacement (AVR) remains a clinical challenge in cardiac surgery. Continuous suture technique (CST) has been considered to allow implantation of relatively larger prostheses compared with interrupted pledgeted suture (IPS). This study aimed to compare early clinical outcomes of CST versus IPS in patients with SAA who underwent surgical AVR.

Methods

Between 2023 to 2025, 250 patients who received surgical AVR with implanted prostheses size <23 mm at our department were enrolled for retrospective analysis: 99 in the CST group and 151 in the IPS group. Early clinical outcomes were analyzed using propensity score matching, with postoperative peak aortic velocity as the primary endpoint.

Results

The CST group exhibited comparable early clinical outcomes to the IPS group, with no postoperative paravalvular leak observed. Shorter surgery time (P=0.03) and cross-clamp time (P=0.045) were shown in the raw CST group; this advantage also reached a significant difference in the isolated AVR subgroup analysis. Our cohort presented that AVR with CST was associated with significantly lower peak aortic velocity compared with IPS in both raw cohort {2.0 [interquartile range (IQR), 1.7–2.2] vs. 2.2 (IQR, 2.0–2.6) m/s; P<0.001} and propensity-matched analyses (P=0.03). Multivariable linear regression also confirmed CST as an independent predictor of reduced postoperative peak aortic velocity (coefficient −0.192; 95% confidence interval, −0.292 to −0.092; P<0.001).

Conclusions

The treatment with CST in AVR could be a valid and well-promoted alternative strategy for patients with SAA and showed a trend toward better hemodynamic performance compared to the traditional IPS approach.

Highlight box.

Key findings

• Continue suture technique (CST) in aortic valve replacement showed a trend toward shorter operative time and significantly lower peak aortic velocity compared with interrupted pledgeted suture (IPS) for small aortic annulus (SAA) patients.

What is known and what is new?

• Aortic root enlargement is routinely performed for SAA patients, which may carry increased surgical complexity and higher risk of operative mortality and complications. CST has been considered to allow implantation of larger prostheses compared with IPS, while clinical evidence supporting its efficacy remains limited.

• Using CST in aortic valve replacement achieved comparable early outcomes with shorter operative time and it also provided significantly lower peak aortic velocity and peak pressure gradient than the traditional IPS.

What is the implication, and what should change now?

• CST in aortic valve replacement could be a valid alternative strategy for SAA patients, offering better hemodynamic performance compared to traditional IPS. Longer-term follow-up studies are necessary to determine whether the hemodynamic advantage in CST is maintained over time.

Introduction

The presence of a small aortic annulus (SAA) in patients undergoing aortic valve replacement (AVR) remains a clinical challenge in cardiac surgery. A clear consensus regarding the cutoff value for defining SAA remains lacking. In surgical practice, SAA is most commonly defined as an annulus unable to accommodate a prosthesis >21 mm, or an aortic annulus ≤23 mm measured by echocardiography or direct intraoperative sizing (1). SAA patients have been reportedly associated with an increased risk of prosthesis-patient mismatch due to the suboptimal valve hemodynamics, which may result in major adverse cardiovascular events, including left ventricular outflow tract obstruction, elevated transvalvular pressure gradients, congestive heart failure, and increased overall mortality after AVR (2,3).

Surgeons routinely perform aortic root enlargement procedures to implant larger prostheses in SAA patients to prevent prosthesis-patient mismatch; other surgical strategies include Ross procedure, supra-annular valves, and stentless bioprostheses (1,4). These options, however, may involve technical complexity, prolong operative time, and increase complication rates such as reoperation for bleeding and pacemaker implantation (5). On the other hand, compared to the traditional interrupted pledgeted suture (IPS) approach, the continuous suture technique (CST) has been considered to provide implantation of a larger prosthesis within the aortic annulus (6,7), which may become an alternative strategy for SAA patients. Suturing techniques differ largely according to surgeon’s individual preference and patient’s anatomy, and the interrupted pledget-reinforced mattress sutures remain the mainstream. Although CST has been associated with shorter operation duration, concerns persist regarding its risk of paravalvular leak (8), and there is a scarcity of reports on its efficacy. Therefore, in this study, we aimed to compare the early outcomes associated with the application of CST versus IPS in AVR for SAA patients. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2001/rc).

Methods

Patients

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. All data in the cohort were approved for research use by the ethics review committee of Guangdong Provincial People’s Hospital (No. XJS2024-007-01; February 2, 2024). Given the retrospective nature of this study, the requirement for individual informed consent was waived.

In this study, we enrolled consecutive patients who underwent surgical AVR with <23 mm prostheses at our department from January 2023 to January 2025. Exclusion criteria included: patients with concomitant aortic arch replacement, Bentall procedure, or aortic annular enlargement procedure; previous AVR; and missing prosthesis size data (Figure 1). A total of 250 patients were eligible for the final analysis, including 99 in the CST group and 151 in the IPS group. Clinical data from laboratory test reports, radiological examination reports, and medical charts were obtained.

Figure 1.

Study flowchart.

Postoperative transthoracic echocardiographic exam was routinely performed before discharge. Given that few patients exhibited a postoperative elevated peak aortic velocity (Vmax) >3.0 m/s and the effective orifice area measurements were not available for all patients (9), we selected Vmax as the primary endpoint of our study. Secondary endpoints included the duration of surgery time, cardiopulmonary bypass (CPB) time, cross-clamp time, intensive care unit stay, postoperative in-hospital stay, peak transvalvular gradient, ejection fraction, paravalvular leak, 30-day mortality, intra-aortic balloon pump support, extracorporeal membrane oxygenation support, stroke, newly required dialysis, and ventilation time >72 h.

Surgical technique

After median sternotomy, CPB was initiated via aortic and right atrial cannulation. An aortic root vent and a left ventricular vent inserted through the right superior pulmonary vein were placed. With the aorta cross-clamped, cardioplegic solution was administered in antegrade fashion or directly into the coronary ostia for patients with aortic insufficiency. After careful resection of the diseased aortic valve and decalcification of the annulus, the aortic annulus diameter was measured using sizers to select a prosthesis of appropriate size.

For AVR using CST, three 2-0 polypropylene sutures were placed in each sinus. We routinely commenced suture at the right-noncoronary commissure, in a “counter-clockwise” direction toward the left-noncoronary commissure, with the needle passing from the aortic side of the annulus, toward the ventricular side, and then through the prosthetic valve sewing ring from bottom to top. Typically, six stitches were made between two commissures (Figure 2). Some may apply a temporary silk suture to gathered outer loops of each parachute suture to minimize suture entanglement and looping. The prosthetic valve was parachuted into the annulus with sutures pulling on and tightening by a nerve hook. Then direct inspection through open leaflets ensured the elimination of redundant suture loops, followed by tying the knots behind the commissures. In the IPS group, valve replacements were performed with pledgeted 2-0 double-ended needle polyester stitch (Ethicon, Inc., Somerville, NJ, USA) for intermittent mattress sutures, with pledgets positioned above the aortic annulus. Finally, the aortotomy was closed by routine procedures. A transesophageal echocardiographic evaluation was performed before surgery and after weaning from the CPB for all cases.

Figure 2.

Continuous suture technique for aortic valve replacement. Suturing was stared at the right-noncoronary commissure, in a “counter-clockwise” direction toward the left-noncoronary commissure, with needle passing from the aortic side of the annulus, toward the ventricular side, and then through the prosthetic valve sewing ring from bottom to top.

The procedure was an isolated AVR in 100 patients (30 in the CST group and 70 in the IPS group), concomitant mitral valve replacement in 120 (58 in the CST group and 62 in the IPS group); other procedures performed simultaneously included coronary artery bypass grafting in 11, Wheat procedure in 21, and tricuspid valvuloplasty in 106. Prosthesis brands are detailed in Figure 3.

Figure 3.

Distribution of prosthesis brands. Comparison of the manufacturers of prosthetic valves used in the IPS and the CST groups. CST, continuous suture technique; IPS, interrupted pledged suture.

Statistical analyses

All analyses were conducted in R version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables are summarized as mean ± standard deviation for normally distributed data or median with an interquartile range (IQR) for non-normal distributions, and were compared using one-way analysis of variance or Kruskal-Wallis test, respectively. Categorical variables are expressed as frequencies with percentages and were analyzed using the chi-square test, or Fisher’s exact test for small sample sizes (n<5).

A propensity score matching model was used to approximate a randomized trial with a 1:1 matched pair design (caliper of width equal to 0.1) through the MatchIt package of R, including matching covariates: age, gender, body surface area, diabetes, coronary artery disease, cerebrovascular disease, previous cardiac surgery, bicuspid aortic valve, ejection fraction, moderate to severe aortic stenosis/aortic insufficiency, prosthesis size (as a categorical variable) and type. Standardized mean difference was assessed for the goodness-of-fit: ≤10% was considered ideal (10). Differences of variables between two matched cohorts were analyzed by paired t-test or Wilcoxon signed rank test for continuous data, and McNemar test for categorical data. A multivariable linear regression model was constructed to estimate independent predictors of the postoperative Vmax. Variables were selected based on univariate screening (P<0.10) and clinical experience. This model specifically assessed the effect of CST on the postoperative Vmax. A two-sided P<0.05 was considered to indicate statistical significance.

Results

Patient demographics

Patients in the overall raw cohort had a mean age of 56±10 years, with 41.2% (n=103) male, 31.6% (n=79) bicuspid aortic valve, and 6.8% (n=17) infective endocarditis. Besides, mechanical valves (76.0%) and 21 mm-sized prostheses (52.4%) accounted for the largest proportion of prosthesis type and size, respectively. As shown in Table 1, significantly higher percentages of male patients (47.7% vs. 31.3%; P=0.02) and moderate to severe aortic insufficiency (31.1% vs. 15.2%; P=0.007) were exhibited in the IPS group compared with the CST group. Meanwhile, the proportion of prosthesis size (P=0.02) and mechanical valve usage (IPS vs. CST: 68.9% vs. 86.9%; P=0.002) differed significantly between the two groups.

Table 1. Preoperative characteristics and prosthesis information before and after propensity score matching.

Variables	Raw				Matched
	IPS (n=151)	CST (n=99)	P value		IPS (n=74)	CST (n=74)	SMD
Age (years)	56±11	56±10	0.71		55±8	56±10	0.02
Male	72 (47.7)	31 (31.3)	0.02		31 (41.9)	28 (37.8)	0.08
BSA (m2)	1.6±0.2	1.6±0.2	0.09		1.6±0.2	1.6±0.2	0.02
Diabetes	13 (8.6)	16 (16.2)	0.11		8 (10.8)	7 (9.5)	0.045
Coronary artery disease	13 (8.6)	6 (6.1)	0.62		5 (6.8)	5 (6.8)	<0.001
Cerebrovascular accident	8 (5.3)	9 (9.1)	0.36		5 (6.8)	4 (5.4)	0.057
COPD	1 (0.7)	0	>0.99		0	0	<0.001
Previous cardiac surgery	4 (2.6)	0	0.26		0	0	<0.001
Infective endocarditis	12 (7.9)	5 (5.1)	0.53		4 (5.4)	4 (5.4)	<0.001
Bicuspid aortic valve	55 (36.4)	24 (24.2)	0.059		24 (32.4)	20 (27.0)	0.12
Echocardiographic data
EF (%)	64 [60–66]	63 [60–66]	0.85		64 [60–66]	63 [60–66]	0.04
EF <50%	10 (6.6)	5 (5.1)	0.81		3 (4.1)	3 (4.1)	<0.001
Moderate to severe AS	70 (46.4)	45 (45.5)	0.99		39 (52.7)	36 (48.6)	0.08
Moderate to severe AI	47 (31.1)	15 (15.2)	0.007		13 (17.6)	13 (17.6)	<0.001
Abnormal wall motion	2 (1.3)	1 (1.0)	>0.99		2 (2.7)	1 (1.4)	0.10
Prosthesis size			0.02				0.03
19	38 (25.2)	15 (15.2)			13 (17.6)	13 (17.6)
20	9 (6.0)	12 (12.1)			7 (9.5)	7 (9.5)
21	83 (55.0)	48 (48.5)			36 (48.6)	35 (47.3)
22	21 (13.9)	24 (24.2)			18 (24.3)	19 (25.7)
Mechanical valve	104 (68.9)	86 (86.9)	0.002		63 (85.1)	61 (82.4)	0.07

Data are presented as n (%), mean ± SD or median [IQR]. AI, aortic insufficiency; AS, aortic stenosis; BSA, body surface area; COPD, chronic obstructive pulmonary disease; CST, continuous suture technique; EF, ejection fraction; IPS, interrupted pledgeted suture; IQR, interquartile range; SD, standard deviation; SMD, standardized mean difference.

After matching, there was a good balance in terms of preoperative patient demographics and prosthesis-related information (Table 1). The standardized mean differences between the two groups are displayed as love plots in Figure 4.

Figure 4.

Love plot for standardized mean differences between the continuous suture technique and the interrupted pledgeted suture groups before and after propensity score matching. AI, aortic insufficiency; AS, aortic stenosis; COPD, chronic obstructive pulmonary disease; CST, continuous suture technique; IPS, interrupted pledgeted suture.

Early postoperative outcomes

Patients undergoing concomitant coronary artery bypass grafting, Wheat procedure, mitral valve replacement, mitral valvuloplasty, and tricuspid valvuloplasty in the overall cohort accounted for 4.4%, 8.4%, 48.0%, 8.0%, 42.4%, respectively. With the exception of a higher mitral valve replacement rate in the raw CST group (58.6% vs. 41.1%, P=0.01), no other difference was found in concomitant surgery between the IPS and the CST groups before and after propensity score matching (Table 2). Surgery time [230 (IQR, 194–274) vs. 250 (IQR, 216–285) min; P=0.03] and cross-clamp time [92 (IQR, 77–118) vs. 102 (IQR, 82–123) min; P=0.045] were significantly shorter in the CST group compared to the IPS group. After matching, the surgery time (P=0.36), CPB time (P=0.64), and cross-clamp time (P=0.24) were comparable between groups. Notably, for the isolated AVR subgroup, significant reductions in surgery time (P=0.005), CPB time (P=0.002), and aortic cross-clamp time (P<0.001) were observed in the CST group versus the IPS group (Table 3).

Table 2. Intraoperative and postoperative outcomes before and after propensity score matching.

Variables	Raw				Matched
	IPS (n=151)	CST (n=99)	P value		IPS (n=74)	CST (n=74)	P value
Combined surgery
CABG	8 (5.3)	3 (3.0)	0.59		4 (5.4)	2 (2.7)	0.68
Wheat procedure	10 (6.6)	11 (11.1)	0.31		5 (6.8)	9 (12.2)	0.42
Mitral valve replacement	62 (41.1)	58 (58.6)	0.01		34 (45.9)	41 (55.4)	0.31
Mitral valvuloplasty	14 (9.3)	6 (6.1)	0.50		2 (2.7)	6 (8.1)	0.29
Tricuspid valvuloplasty	59 (39.1)	47 (47.5)	0.24		27 (36.5)	32 (43.2)	0.50
Surgery time (min)	250 [216–285]	230 [194–274]	0.03		247 [215–280]	238 [196–273]	0.36
CPB time (min)	153 [118–180]	137 [107–184]	0.12		150 [117–179]	141 [108–184]	0.64
Cross-clamp time (min)	102 [82–123]	92 [77–118]	0.045		102 [82–123]	95 [77–117]	0.24
Early outcomes
30-day mortality	3 (2.0)	1 (1.0)	0.93		1 (1.4)	1 (1.4)	>0.99
IABP support	3 (2.0)	2 (2.0)	>0.99		1 (1.4)	0	>0.99
ECMO support	3 (2.0)	0	0.41		1 (1.4)	0	>0.99
Stroke	2 (1.3)	0	0.67		0	0	>0.99
Newly required dialysis	5 (3.3)	0	0.17		1 (1.4)	0	>0.99
Ventilation >72 h	16 (10.6)	4 (4.0)	0.10		7 (9.5)	2 (2.7)	0.18
ICU stay (days)	2 [1–3]	2 [1–3]	0.008		2 [1–3]	2 [1–3]	0.09
Postoperative in-hospital stay (days)	7 [5–8]	6 [5–7]	0.29		6 [5–8]	6 [5–7]	0.10
Echocardiographic data
Paravalvular leak	0	0	>0.99		0	0	>0.99
Peak aortic velocity (m/s)	2.2 [2.0–2.6]	2.0 [1.7–2.2]	<0.001		2.2 [1.9–2.5]	2.0 [1.8–2.2]	0.003
PPG (mmHg)	21 [17–27]	15 [12–19]	<0.001		19 [15–24]	16 [14–20]	0.004
LVEDD (mm)	45 [41–48]	44 [40–47]	0.12		45 [42–47]	44 [40–47]	0.33
LVESD (mm)	29 [26–32]	29 [26–32]	0.61		29 [26–32]	30 [26–33]	0.97
EF (%)	62 [57–65]	60 [57–64]	0.61		62 [57–65]	60 [57–64]	0.66
EF <50%	16 (10.6)	7 (7.1)	0.47		6 (8.1)	6 (8.1)	>0.99

Data are presented as median [interquartile range] or n (%). CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass; CST, continuous suture technique; ECMO, extracorporeal membrane oxygenation; EF, ejection fraction; IABP, intra-aortic balloon pump; ICU, intensive care unit; IPS, interrupted pledgeted suture; LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; PPG, peak pressure gradient.

Table 3. Surgery duration in subgroup.

Variables	Isolated AVR		P value
	IPS (n=70)	CST (n=30)
Combined surgery
CABG	0	0	>0.99
Wheat procedure	0	0	>0.99
Mitral valve replacement	0	0	>0.99
Mitral valvuloplasty	12 (17.1)	6 (20.0)	0.96
Tricuspid valvuloplasty	9 (12.9)	3 (10.0)	0.95
Surgery time (min)	206 [188–247]	175 [158–227]	0.005
CPB time (min)	115 [98–146]	86 [75–117]	0.002
Cross-clamp time (min)	80 [65–100]	55 [47–79]	<0.001

Data are presented as median [interquartile range] or n (%). AVR, aortic valve replacement, CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass; CST, continuous suture technique; IPS, interrupted pledgeted suture; IQR, interquartile range.

There was no difference in terms of 30-day mortality (2.0% vs. 1.0%; P=0.93), intra-aortic balloon pump support (2.0% vs. 2.0%; P>0.99), extracorporeal membrane oxygenation support (2.0% vs. 0; P=0.41), stroke (1.3% vs. 0; P=0.67), newly required dialysis (3.3% vs. 0; P=0.17), and prolonged ventilation (10.6% vs. 4.0%; P=0.10) between the IPS and the CST groups (Table 2). The intensive care unit stay was shorter for patients with CST (P=0.008); the postoperative in-hospital stay (P=0.29) was equivalent between groups. Similar early outcomes were demonstrated after matching, with no significant differences in 30-day mortality, other adverse events, length of intensive care unit stay and postoperative in-hospital stay.

Postoperative echocardiographic data

No substantial paravalvular leak was found on intraoperative and postoperative echocardiography in either group. The Vmax [2.0 (IQR, 1.7–2.2) vs. 2.2 (IQR, 2.0–2.6) m/s; P<0.001] and peak pressure gradient [15 (IQR, 12–19) vs. 21 (IQR, 17–27) mmHg; P<0.001] were lower for patients undergoing CST than those in the IPS group (Table 2). Similar outcomes were demonstrated after matching (Vmax: P=0.003; peak pressure gradient: P=0.004). No significant difference was observed in left ventricular end-diastolic diameter (raw: P=0.12; matched: P=0.33), left ventricular end-systolic diameter (raw: P=0.61; matched: P=0.97), ejection fraction (raw: P=0.61; matched: P=0.66), and ejection fraction <50% (raw: P=0.47; matched: P>0.99) in either raw or matched cohort.

Independent predictors of postoperative Vmax

The multivariable linear regression model (R²=0.33, adjusted R²=0.30, P<0.001; Table 4) revealed that CST was associated with lower postoperative Vmax [coefficient −0.192; 95% confidence interval (CI), −0.292 to −0.092; P<0.001] compared to IPS approach. Moreover, body surface area (coefficient 0.661; 95% CI, 0.300 to 1.023; P<0.001), prosthesis size (coefficient −0.153; 95% CI, −0.205 to −0.101; P<0.001), and bioprosthetic valve (coefficient 0.258; 95% CI, 0.116 to 0.399; P<0.001) were indicated as the independent predictors of postoperative Vmax.

Table 4. Multivariable linear regression analyses for peak aortic velocity after aortic valve replacement.

Variables	Coefficient (95% CI)	P value
Continuous suture technique	−0.192 (−0.292, −0.092)	<0.001
BSA	0.661 (0.300, 1.023)	<0.001
Prosthesis size	−0.153 (−0.205, −0.101)	<0.001
Bioprosthetic valve	0.258 (0.116, 0.399)	<0.001

The following variables were included in the multivariable linear regression model [suture technique, age, gender, BSA, valve type, prosthesis size (as a continuous variable), previous cardiac surgery, concomitant mitral valve replacement, concomitant Wheat procedure, ejection fraction <50% and bicuspid aortic valve], with model R square (0.33) and adjusted R square (0.30). The model was checked for multicollinearity with the variance inflation factor, for which all values were <2. BSA, body surface area; CI, confidence interval.

Discussion

This study demonstrates that CST achieved considerable early outcomes compared with the traditional IPS approach in SAA patients undergoing AVR. There was no postoperative paravalvular leak in our cohort and shorter surgery durations were revealed for the CST group in the isolated AVR subgroup analysis. In our practical experience, AVR with CST provided a larger prosthetic valve size than the IPS approach, which may lead to superior hemodynamic performance. This theoretical advantage was corroborated by our findings, which demonstrated that CST presented significantly lower postoperative Vmax and peak pressure gradient compared to the traditional IPS. Moreover, both the propensity score matching method and multivariable linear regression substantiated this foregoing result. Our results illustrated that AVR with CST could be safe, effective, and well promoted for SAA patients and showed a trend toward better hemodynamic performance compared to the traditional IPS approach.

Limited data are available considering the prevalence of SAA patients. The definition of SAA varies in different research, including criteria based on prosthesis size, body surface area (11), and aortic sinotubular junction diameter indexed for body height (12). Given that 23 mm prosthesis was the most used in clinical practice, our cohort enrolled patients receiving prosthesis sized <23 mm for analysis. Also, our experience revealed that in cases where a 21-mm Regent sizer represented the maximum size able to pass through the annulus, yet CST allowed the implantation of a 22 mm AP360 prosthesis. Accordingly, we included patients with a measured annulus diameter of 21 mm who received 22 mm prostheses (all of which in this cohort were AP360 model) for analysis. It was presented that females accounted for a relatively higher proportion among SAA patients in our cohort, with this sex preponderance becoming stronger as prosthesis size decreased (Figure 5), which was in line with other studies (1,13). Besides, bicuspid aortic valve disease was present in 1/3 of SAA patients, a prevalence higher than that reported in the general population (0.5–2%) (14). Whether an association exists among SAA, general, and bicuspid aortic valve requires further exploration.

Figure 5.

Distribution of patient gender stratified by prosthetic valve size.

Aortic root enlargement is widely regarded as the primary option for SAA patients undergoing surgical AVR, with precise advantages focusing on a lower incidence of prosthesis-patient mismatch and a larger indexed effective valve orifice area in comparison with conventional AVR (1,15). However, its increased surgical complexity leads to prolonged surgical times, and some studies reported a higher operative mortality and other adverse events (4,16). AVR with CST, using monofilament suture and freedom from pledgeted material, has emerged as an effective alternative approach for SAA. It allows the implanted prosthetic valve to be approximately 1- or 2-size larger compared to the IPS demonstrated both in our clinical practice and other studies (6,8). The benefit of CST has been primarily illustrated in previous studies as shorter CPB and cross-clamp times (17). While no significant time-saving benefit was observed in CST of our cohort, which might be likely attributed to the high proportion of concomitant procedures performed and the limited sample size. The isolated AVR subgroup analysis substantiated this advantage.

Another finding to be highlighted in our study was better hemodynamics (lower Vmax) achieved in CST for SAA patients undergoing AVR compared with the traditional IPS. Larger prosthetic valve size enabled by CST may lead to better effective orifice area and promote hemodynamic variables. Rather than being wedged with pledgeted material, AVR with CST let the prosthesis directly seat on the aortic annulus. Also, CST prevents suture-induced annular reduction in the interrupted suturing during leaflet positioning and knot tying. A study based on combined experimental and computational tools showed that the pledgetted sutured biological prosthetic valves caused flow disturbances, which in turn increased the mean pressure gradient and decreased the effective orifice area (18). A recent prospective multicenter trial with a propensity score-matched method concluded that, although clinical outcomes were comparable between AVR with pledgeted and nonpledgeted sutures, pledgets might lead to a slightly smaller effective orifice area in the long run (19). Longer-term follow-up studies are necessary to determine whether the above hemodynamic advantage in CST is maintained over time. In addition, by minimizing the suture gap, perivalvular knots, and pledgets, CST may decrease potential accumulation of bacteria and fibrin clots, which may consequently lower the risks of thrombogenesis, infective endocarditis, and prosthetic valve obstruction (20).

Previous studies questioned the increased risk of paravalvular leak and reoperation rate associated with CST for patients undergoing AVR (8). Moreover, some have cautioned against its use in patients with infective endocarditis. A study based on a 10-year follow-up reported a 12% incidence of moderate or severe paravalvular leak after mechanical AVR using semicontinuous suture, compared to 0% in IPS, which also suggested that the risk of developing paravalvular leak persisted beyond the first year after replacement (8). In contrast, Bardia et al. recently demonstrated that CST was not associated with paravalvular leak for stented bioprosthetic AVR, even though their data included patients with higher rate of preoperative infective endocarditis in the CST group (2.9% vs. 0.6%, P=0.001) (17). They proposed that the higher paravalvular leak rates in CST were attributed to the inappropriate flat-shaped “annular” sewing ring in mechanical valves. In our cohort, no paravalvular leak was observed in the early postoperative period for either the CST or the IPS group. This outcome likely reflected the importance of careful manipulation of the aortic annulus, appropriate distance, and even stitching.

Body surface area and prosthesis size have been confirmed as significant factors influencing blood velocity across the prosthetic valve (9), which provided a compelling explanation for our multivariable linear regression findings. Interestingly, bioprosthetic valves also showed a trend toward higher postoperative Vmax in our analyses. This observation aligned with a recent meta-analysis revealing a statistically significantly higher incidence of prosthesis-patient mismatch after AVR with bioprosthetic valves versus mechanical valves (21). Future studies are needed to determine the association between prosthesis type and hemodynamic gradients.

Limitation

Several limitations should be acknowledged. First, this study is likely subject to selective bias and retrospective design. Although a propensity score matching model was used for adjustment, the surgeon’s preference and uncollected clinical data may influence the results. The lack of postoperative effective orifice area also prevented the identification of patient-prosthesis mismatch and limited a more direct outcome analysis. Also, the analysis was challenged by the relatively small sample size and its limitation to a single center’s experience. A comparative analysis with a larger dataset—for instance, by including patients with prosthesis size ≥23 mm—would better highlight the advantage of the CST in SAA patients. A longer follow-up is required to determine the safety and efficacy of CST.

Conclusions

In SAA patients undergoing AVR, CST achieved comparable early outcomes to the traditional IPS approach in our cohort, and significantly lower Vmax and peak pressure gradient were shown in the CST group. These findings suggest that CST in AVR showed a trend toward better hemodynamic performance compared to IPS, which could be a valid and well-promoted alternative strategy for SAA patients.

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. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. All data in the cohort were approved for research use by the ethics review committee of Guangdong Provincial People’s Hospital (No. XJS2024-007-01; February 2, 2024). Given the retrospective nature of this study, the requirement for individual informed consent was waived.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2001/dss

jtd-18-01-5-dss.pdf ^{(67.9KB, pdf)}

DOI: 10.21037/jtd-2025-aw-2001