Comparison of recent clinical effects between right anterior mini- thoracotomy and full sternotomy used in Bentall surgery

Abstract

Background

The Bentall procedure remains the gold standard for treating aortic root pathologies. However, the conventional full sternotomy (FS) approach is associated with sternal complications and prolonged recovery times. The right anterior mini-thoracotomy (RAMT), a minimally invasive alternative, offers reduced trauma and preserves sternal integrity, but its safety and efficacy require further evaluation. This study aims to compare the short-term outcomes of RAMT and FS in Bentall procedures.

Methods

This single-center retrospective study included all patients who underwent Bentall surgery between May 2022 and December 2024. Using a 1:2 matched design to balance baseline characteristics, 16 patients were included in the RAMT group and 32 in the FS group. The primary endpoint was the 30-day incidence of major adverse cardiac and cerebrovascular events (MACCEs). Secondary endpoints included cardiopulmonary bypass time (CPBT), postoperative complications, and short-term follow-up outcomes.

Results

The aortic cross-clamp time was significantly longer in the RAMT group compared to the FS group (146±22.4 vs. 130±27.9 min, P=0.03), though no significant differences were found in CPBT or total operative time. Intraoperative blood loss was significantly lower in the RAMT group (median 300 vs. 500 mL, P=0.002). No significant differences were observed between the two groups in 30-day MACCEs, mechanical ventilation time, postoperative length of stay, total hospitalization expenditures, or incidence of major postoperative complications. Short-term follow-up (median 12 months) showed that all patients, except for one case of prosthetic valve endocarditis (PVE) in the RAMT group, survived with improved cardiac function.

Conclusions

In patients selected according to strict criteria, the RAMT approach for Bentall surgery was associated with longer aortic cross-clamp time but significantly reduced intraoperative blood loss, while demonstrating comparable perioperative safety and short-term efficacy to the conventional FS approach. RAMT is a feasible minimally invasive option, although its long-term outcomes warrant further validation through larger sample sizes and extended follow-up.

Highlight box.

Key findings

• This retrospective study found that the right anterior mini-thoracotomy (RAMT) approach for Bentall surgery provided comparable safety and short-term efficacy to full sternotomy (FS). While aortic cross-clamp time was longer with RAMT, it significantly reduced intraoperative blood loss and completely avoided sternal complications.

What is known and what is new?

• FS is the traditional approach for Bentall procedures but carries risks of sternal complications and significant bleeding.

• The RAMT approach for Bentall surgery achieves comparable perioperative outcomes while significantly reducing intraoperative blood loss and avoiding sternal complications. This study provides objective evidence of the technique's learning challenges and confirms its clinical feasibility as a minimally invasive alternative for aortic root replacement.

What is the implication, and what should change now?

• This study provides clinical evidence supporting RAMT as a feasible minimally invasive option for the Bentall procedure, although its long-term efficacy requires further validation through larger sample sizes and extended follow-up periods.

Introduction

Since its introduction by Bentall and De Bono in 1968, composite valve-graft conduit replacement (the Bentall procedure) has been the standard treatment for aortic root pathologies including aneurysms, acute type A aortic dissection, and Marfan syndrome (1,2). The conventional FS provides excellent exposure but is associated with several risks, including sternal dehiscence (1–5%), mediastinitis (0.5–5%), chronic pain, and respiratory impairment, all of which can significantly delay recovery (3-5). These drawbacks have led to growing interest in less invasive alternatives that reduce surgical trauma and align with Enhanced Recovery After Surgery (ERAS) principles (6).

The RAMT approach, performed through a 5–7 cm incision in the 2nd or 3rd right intercostal space with thoracoscopic assistance, preserves sternal integrity and may help mitigate many of the complications associated with FS. Preliminary series, such as that by Ji et al. (n=15), have reported favorable early outcomes with zero mortality and short hospital stays (7). A multicenter study (n=44) further documented no 30-day mortality and drainage volumes comparable to those of FS, without conversion to sternotomy (4). Despite these promising results, the RAMT approach still requires more robust evidence to support its widespread use.

In this context, our study retrospectively analyzed patients who underwent the Bentall procedure at our institution, comparing the perioperative outcomes between the RAMT and FS groups. The primary objective of this study was to compare the perioperative outcomes between the two approaches, using the 30-day incidence of major adverse cardiac and cerebrovascular events (MACCEs) as the primary endpoint. The findings are intended to provide clinical evidence that may inform the adoption of RAMT in minimally invasive aortic root surgery. We present this article in accordance with the STROBE and SUPER reporting checklists (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2216/rc).

Methods

Clinical data and grouping

This retrospective study initially screened all consecutive patients who underwent Bentall procedures at The Second Affiliated Hospital, Zhejiang University School of Medicine between May 2022 and December 2024, and who met the standard surgical indications.

Inclusion criteria

(I) First-time cardiac surgery; (II) aortic valve disease (moderate-to-severe aortic regurgitation and/or stenosis) with a sinus of Valsalva diameter ≥50 mm; (III) New York Heart Association (NYHA) functional class III or higher; (IV) no history of right chest wall trauma, surgery, or radiotherapy; (V) no severe malformations of the iliac, abdominal aorta, or femoral arteries.

Exclusion criteria

(I) Emergency surgery, redo cardiac surgery, or concomitant structural heart disease; (II) chest wall deformity, severe pleural adhesions, or extreme obesity [body mass index (BMI) >30 kg/m²]; (III) severe chronic obstructive pulmonary disease, pulmonary fibrosis, or prior lung surgery; (IV) severely impaired cardiac function [left ventricular ejection fraction (LVEF) <30%]; (V) severe hepatic or renal dysfunction or coagulation disorders; (VI) life-threatening arrhythmias (ventricular flutter/fibrillation or ≥ second-degree atrioventricular block); (VII) hemodynamically unstable conditions (progressive hypotension, shock, acute heart failure, myocardial infarction, or altered mental status).

Matching and grouping

The selection of the surgical approach was based on strict multidisciplinary evaluation criteria, which included anatomical characteristics, cardiopulmonary function, and patient preference. The final decision was made after a comprehensive discussion. Patients were then divided into the RAMT group and the FS group based on the surgical approach.

Given the limited sample size in the RAMT group (n=16), a multi-step matching protocol was implemented. First, two attending physicians with over 10 years of clinical experience independently screened all potential control candidates from the electronic medical records of patients who underwent FS between May 2022 and December 2024, according to predefined inclusion and exclusion criteria. Key matching variables were predetermined based on clinical relevance, including demographic characteristics (sex, age), baseline comorbidities (hypertension, diabetes, hyperlipidemia, renal function staging), and cardiac function parameters (ejection fraction, left ventricular end-diastolic diameter). Age and sex were prioritized as strong confounding factors. Ultimately, 32 matched controls were selected from 42 candidates, forming a 16:32 matched cohort. The balance of baseline characteristics before and after matching was assessed using chi-square tests, Wilcoxon rank-sum tests, and independent t-tests, with standardized mean differences (SMDs) calculated.

To explore potential trends related to accumulating surgical experience, the RAMT cases were divided into two sequential groups: the first 7 cases (earlier phase) and the subsequent 9 cases (later phase). Operative time, cardiopulmonary bypass time (CPBT), and aortic cross-clamp time were compared between these two groups in an exploratory manner.

Outcome measures

Perioperative data were collected for both patient groups. The primary endpoint was the 30-day composite incidence of MACCEs, including all-cause mortality, acute myocardial infarction, cerebral infarction. Key secondary endpoints included prosthetic valve endocarditis (PVE), and re-exploration for bleeding or cardiac tamponade, CPBT, prosthetic valve regurgitation, and major postoperative complications, such as arrhythmia, acute kidney injury, and pericardial effusion. The data completeness for key variables was very high in this study. For the very few instances of missing follow-up data, they were treated as missing values in the analysis and handled using median imputation.

Surgical technique

Implanted devices and equipment

The prosthetic valves used in this study included bioprosthetic valves (Carpentier-Edwards PERIMOUNT, California, USA; Bairensi, Beijing, China) and mechanical valves (St. Jude Medical Regent, Minnesota, USA). Vascular grafts were supplied by MAQUET (Rastatt, Germany) products. The left ventricular assist device used was the Impella P50, an investigational device (Anhui Tongling Fangsheng Technology Co., Ltd., Tongling City, China).

RAMT group

Patient positioning and preparation

(I) The patient was placed in the supine position with the right chest elevated 30°. (II) General anesthesia with double-lumen endotracheal intubation was performed, ensuring right lung collapse. (III) Standard monitoring included ECG, invasive arterial/venous pressure, and cardiac output. (IV) External defibrillator pads were applied.

Surgical approach (Figure 1)

Figure 1.

Bentall surgery procedure. (A) Right anterior mini-incision; (B) mobilization of the coronary arteries; (C) perform valve sizing using a sizer; (D) implantation of a valved conduit; (E) coronary artery anastomosis; (F) aortic distal anastomosis. This image is published with the participants’ consent.

(I) Femoral cannulation: after routine disinfection and draping, the chest and bilateral groin areas were exposed. A groin incision was made to isolate the femoral artery and vein. Following systemic heparinization, peripheral cardiopulmonary bypass (CPB) was established through the femoral artery and vein. (II) Thoracic access: a direct incision of 6–8 cm was made through the second or third intercostal space, adjacent to the sternum. The subcutaneous tissue, muscles, and pleura were incised layer by layer to access the thoracic cavity. An intercostal retractor was used to expose the surgical field. The fat was dissected, and the pericardium was incised and fully suspended to expose the aortic root. An incision protector was inserted to protect the incision and lift the pericardium upward, bringing the aorta closer to the chest wall. A catheter was inserted through the right superior pulmonary vein for left-heart drainage. (III) CPB management: the patient was cooled during CPB, and carbon dioxide was continuously blown into the surgical area. (IV) Aortic cross-clamping: the ascending aorta was cross-clamped and incised transversely. Under direct vision, cardioplegic solution was anterogradely perfused through the left and right coronary ostia. (V) Root replacement: after satisfactory cessation of the heart’s function, the aortic root was dissected to the level of the aortic valve. The left and right coronary ostia were dissected into “button” shapes. The diseased aortic valve leaflets and part of the ascending aorta were resected. Interrupted mattress sutures with pledgets were placed above the aortic valve annulus. A prosthetic aortic valve conduit was inserted and seated, and the sutures were tied. The valve’s opening and closing functions were tested to ensure normal operation. A 3-0 Prolene suture was used to reinforce the sutures around the aortic valve annulus. A circular hole of 5–10 mm in diameter was made at the appropriate position on the prosthetic vascular graft with an electrocautery pen. A 6-0 Prolene suture was used for continuous suturing to transplant the left and right coronary buttons onto the prosthetic vascular graft. Cold-blood cardioplegic solution was perfused through the graft to assess for bleeding at the anastomotic sites of the left and right coronary ostia. The prosthetic vascular graft was trimmed and anastomosed end-to-end with the distal end of the native aorta. (VI) Weaning from CPB: the patient was rewarmed and placed in the Trendelenburg position. The left heart and aortic root were fully vented. The ascending aorta was unclamped, and the heart resumed beating spontaneously or after defibrillation. Parallel circulation was established, and the dosage of vasoactive drugs was adjusted. The flow rate of CPB was gradually reduced until bypass was discontinued. Protamine was administered to neutralize heparin, and the femoral artery and vein catheters were removed. (VII) Closure: hemostasis was thoroughly achieved. The pericardial cavity was fully opened, a right thoracic drainage tube was placed, and the chest and groin incisions were closed layer by layer.

FS group

The patient was placed in the supine position after general anesthesia. The sternum was split in the midline and spread laterally, and the pericardium was opened and suspended. Extracorporeal circulation was established through the ascending aorta and a single venous line. After clamping, the aortic root was fully dissected, and the coronary ostia were shaped into a “button” configuration. The ascending aorta and aortic valve were resected, and a valve conduit was used for aortic valve replacement. The left and right coronary ostia were grafted onto the artificial vessel, and the distal end was anastomosed end-to-end with the native vessel. After unclamping, CPB was gradually weaned, hemostasis was performed, drainage tubes were placed, the sternum was fixed with wires, and the incision was sutured.

Statistical analysis

Statistical analyses were performed using SPSS 26.0 and R 4.3.3. Continuous variables with a normal distribution were expressed as mean ± standard deviation (SD) and compared using Student’s t-test. Non-normally distributed data were presented as median (interquartile range) and analyzed using the Mann-Whitney U test (Wilcoxon rank-sum test). Categorical variables were reported as numbers (percentages, %) and compared using Pearson’s Chi-squared test or Fisher’s exact test, as appropriate. SMDs were calculated for all baseline variables to assess the balance between groups before and after matching. All tests were two-sided, and a P value <0.05 was considered statistically significant.

Ethical statement

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of The Second Affiliated Hospital, Zhejiang University School of Medicine (No. 2025-Lunshenyan-0626). Given the retrospective nature of the study and the strict protection of patient privacy, the Ethics Committee waived the requirement for patient informed consent.

Results

Comparison of baseline clinical characteristics between groups

A total of 58 patients undergoing Bentall procedures were enrolled in this study and categorized into either the RAMT group (n=16) or FS group (n=42) based on surgical approach. After 1:2 matching (balanced for baseline characteristics), the final cohort comprised 16 RAMT and 32 FS cases for analysis. Pre-match data showed significantly older age in the RAMT group but comparable other baseline parameters. Post-match analysis confirmed effective control of selection bias, ensuring validity for perioperative outcome comparisons (Table 1).

Table 1. Demographic and preoperative data.

Variables	Before matching					After matching
	RAMT (N=16)	FS (N=42)	Poverall	SMD		RAMT (N=16)	FS (N=32)	Poverall	SMD
Male gender	14 (87.5)	35 (83.3)	>0.99	0.118		14 (87.5)	27 (84.4)	>0.99	0.090
Age (years)	67.5 [64.0; 69.0]	59.0 [49.0; 65.0]	0.004	0.962		67.5 [64.0; 69.0]	62.0 [54.8; 67.2]	0.06	0.563
BMI (kg/m2)	22.8 [19.9; 25.0]	23.3 [21.4; 24.9]	0.53	0.028		22.8 [19.9; 25.0]	23.5 [21.4; 24.9]	0.42	0.030
Hypertension	4 (25.0)	16 (38.1)	0.53	0.285		4 (25.0)	13 (40.6)	0.46	0.337
Diabetes	0	2 (4.76)	>0.99	0.316		0	2 (6.25)	0.55	0.365
Hyperlipidemia	4 (25.0)	8 (19.0)	0.72	0.144		4 (25.0)	6 (18.8)	0.71	0.152
Renal insufficiency			0.75	0.055				0.66	0.078
No	13 (81.2)	32 (76.2)				13 (81.2)	22 (68.8)
CKD2	3 (18.8)	7 (16.7)				3 (18.8)	7 (21.9)
CKD3	0	3 (7.14)				0	3 (9.38)
Cerebral infarction	6 (37.5)	7 (16.7)	0.16	0.482		6 (37.5)	5 (15.6)	0.14	0.511
Atrioventricular block	1 (6.25)	3 (7.14)	>0.99	0.036		1 (6.25)	1 (3.12)	1.00	0.148
Ejection fraction (%)	53.1±11.4	57.2±9.92	0.22	0.385		53.1±11.4	58.6±10.0	0.11	0.512
LVEDD (mm)	6.56 [6.17; 7.39]	6.24 [5.64; 7.05]	0.14	0.418		6.70±0.84	6.25±0.99	0.11	0.478

Data are presented as mean ± SD, median [IQR] or n (%). BMI, body mass index; CKD, chronic kidney disease; FS, full sternotomy; IQR, interquartile range; LVEDD, left ventricular end-diastolic diameter; RAMT, right anterior mini-thoracotomy; SD, standard deviation; SMD, standardized mean difference.

Comparison of perioperative data between the two groups of patients

A comparison of the perioperative data between the two matched groups indicated that RAMT and FS demonstrated generally consistent performance regarding surgical safety and short-term efficacy. The rate of spontaneous return to sinus rhythm recovery was 75.0% (12/16) in the RAMT group and 62.5% (20/32) in the FS group. CPBT was numerically longer in the RAMT group (178±24.7 vs. 165±34.5 min), and aortic cross-clamp time was significantly prolonged in the RAMT group (146±22.4 vs. 130±27.9 min, P=0.03). No statistically significant differences were observed in graft length (47.6±9.06 vs. 50.8±9.57 mm, P=0.34) or intraoperative transfusion volume (median 695 vs. 600 mL). The FS group had significantly greater intraoperative blood loss (P=0.002), with 5 cases requiring hemostatic wrapping (for aortic root-to-right atrial communication) due to difficult bleeding control.

No significant differences were observed in 48-hour postoperative drainage volume (48h-PDV), mechanical ventilation time, postoperative length of stay, echocardiographic parameters such as ejection fraction, aortic valve flow velocity (AVFV), mean pressure gradient (MPG), prosthetic valve competence (assessed by regurgitation and leak), or total hospitalization expenditures. In summary, although RAMT required longer aortic cross-clamp times, the two techniques showed broadly similar safety profiles and short-term outcomes in this matched cohort, with RAMT appearing to offer an advantage in intraoperative hemostasis control (Table 2).

Table 2. Comparison of perioperative data.

Variables	RAMT (N=16)	FS (N=32)	Poverall
Surgical parameters
CPBT (min)	178±24.7	165±34.5	0.12
Cross-clamping time (min)	146±22.4	130±27.9	0.03
Supporting time (min)	19.5 [16.0; 25.5]	25.0 [16.0; 36.2]	0.13
Operation time (min)	275 [235; 324]	288 [245; 321]	0.66
Intraoperative details
Spontaneous return to sinus rhythm	12 (75.0)	20 (62.5)	0.59
Prosthesis type
Biological			0.96
Size 23	0	2 (6.25)
Size 25	8 (50.0)	16 (50.0)
Size 27	3 (18.8)	6 (18.8)
Mechanical			0.59
Size 21	0	1 (3.13)
Size 23	0	1 (3.13)
Size 25	3 (18.8)	3 (9.38)
Size 27	2 (12.5)	3 (9.38)
Vascular			0.62
Size 26	0	2 (6.25)
Size 27	0	1 (3.13)
Size 28	9 (56.2)	12 (37.5)
Size 30	7 (43.8)	17 (53.1)
Graft length (mm)	47.6±9.06	50.8±9.57	0.34
Blood transfusion (mL)	695 [0.00; 1050]	600 [0.00; 1210]	0.86
Blood loss (mL)	300 [275; 300]	500 [300; 500]	0.002
Hemostatic wrapping	0	5 (15.6)	0.15
Echocardiographic data
Ejection fraction (%)	43.3±11.1	49.1±14.2	0.13
AVFV (m/s)	2.20 [1.88; 2.51]	2.00 [1.81; 2.30]	0.16
MPG (mmHg)	9.50 [6.75; 12.0]	8.00 [6.00; 10.2]	0.45
PVR: no	32 (100.0)	16 (100.0)	NA
PVL: no	32 (100.0)	16 (100.0)	NA
LVEDD (mm)	5.49±0.89	5.24±0.78	0.34
Postoperative condition
48h-PDV (mL)	635 [431; 1042]	735 [598; 862]	0.63
Mechanical ventilation time (hours)	23.0 [21.0; 26.2]	20.0 [17.8; 21.2]	0.47
Hospitalization data
Postoperative length of stay (days)	9.50 [8.00; 12.0]	9.50 [8.00; 12.0]	0.79
Expenditures (RMB)	146,637 [128,796; 188,601]	136,925 [123,651; 158,556]	0.43

Data are presented as mean ± SD, median [IQR] or n (%). Supporting time specifically refers to the time interval from the release of the aortic cross-clamp, which restores blood supply to the heart, to the complete weaning from cardiopulmonary bypass. 48h-PDV, 48-hour postoperative drainage volume; AVFV, aortic valve flow velocity; CPBT, cardiopulmonary bypass time; FS, full sternotomy; IQR, interquartile range; LVEDD, left ventricular end-diastolic diameter; MPG, mean pressure gradient; NA, not applicable; PVL, periprosthetic valve leak; PVR, prosthetic valve regurgitation; RAMT, right anterior mini-thoracotomy; SD, standard deviation.

Comparison of postoperative complications between groups

The comparison of postoperative complications between the matched groups (Table 3) revealed that neither the RAMT nor the FS group experienced all-cause mortality and acute myocardial infarction. Regarding specific complications, both groups had case of arrhythmia. Pericardial effusion occurred in 2 cases in the RAMT group and 2 cases in the FS group, while pleural effusion requiring thoracentesis was observed in 1 case in the RAMT group and 4 cases in the FS group. In terms of bleeding management, neither group required re-exploration for bleeding; however, one patient in the FS group required plasma transfusion due to severe intraoperative oozing. Additionally, one patient in the RAMT group developed PVE with vegetation formation, leading to multiple hospital readmissions. In the FS group, notable complications included one patient who developed acute cerebral infarction, another with inadequate wound healing, and one who required temporary support with a domestically produced Impella P50 ventricular assist device for intraoperative low cardiac output syndrome and was successfully weaned. Furthermore, two FS patients required readmission due to pneumonia—one with systemic lupus erythematosus who developed postoperative pneumonia with fever, and another with recurrent aspiration pneumonia that necessitated intensive care unit management.

Table 3. Comparison of postoperative complication data.

Variables	RAMT (N=16)	FS (N=32)	Poverall
Arrhythmia	2 (12.5)	4 (12.5)	>0.99
Dialysis	0	0	NA
Acute myocardial infarction	0	0	NA
Cerebral infarction	0	1 (3.12)	>0.99
Pericardial effusion	2 (12.5)	2 (6.25)	0.59
Thoracentesis	1 (6.25)	4 (12.5)	0.65
RTH	0	0	NA
Inadequate wound healing	0	1 (3.12)	>0.99
Mechanical assist device	0	1 (3.12)	>0.99
PVE	1 (6.25)	0	0.33
Mortality	0	0	NA

Data are presented as n (%). FS, full sternotomy; NA, not applicable; PVE, prosthetic valve endocarditis; RAMT, right anterior mini-thoracotomy; RTH, re-thoracotomy for hemostasis.

Short-term follow-up results

All patients underwent at least two outpatient follow-up visits postoperatively, with a median follow-up duration of 12 months (range, 3–36 months). During the follow-up period, all patients survived and showed significant symptomatic improvement, achieving NYHA functional class I or II. In the RAMT group, one patient developed PVE 3 months postoperatively, with vegetation formation and recurrent fever. Blood cultures identified Staphylococcus capitis subsp. Despite multiple hospital admissions for targeted antibiotic therapy, the infection could not be resolved. The patient subsequently underwent redo sternotomy and is currently recovering well. For all other patients, follow-up echocardiography showed well-functioning prosthetic aortic valves, with no evidence of periprosthetic valve leak (PVL) or moderate/severe aortic regurgitation.

Comparison of different phases in the RAMT group

A phased analysis of surgical data in the RAMT group revealed that with accumulated surgical experience, stage II (later phase, n=9) showed a decreasing trend in CPBT, aortic cross-clamp time, and total operative time compared to stage I (earlier phase, n=7) (see Table 4 and Figure 2).

Table 4. Comparison at different learning stages.

Variables	Stage I (N=7)	Stage II (N=9)	Poverall
CPBT (min)	185±20.9	173±27.2	0.32
Cross-clamping time (min)	153±21.8	141±22.7	0.31
Operative time	303±61.5	268±46.7	0.24

Data are presented as mean ± SD. CPBT, cardiopulmonary bypass time; SD, standard deviation.

Figure 2.

Time trends at different learning stages. Scatter plot of CPBT/cross‑clamp time/operative time for each RAMT case, presented in chronological order. The trend line illustrates the numerical change over time. CPBT, cardiopulmonary bypass time; RAMT, right anterior mini-thoracotomy.

Discussion

In recent years, minimally invasive aortic root surgery techniques have made significant advancements. The primary goals are to accelerate patient recovery by reducing surgical trauma, ensuring procedural safety, and meeting patients’ growing demand for improved cosmetic outcomes. This study focuses on the application of RAMT in Bentall procedures. By integrating practical experience with insights from the literature, we further explore the clinical value and technical challenges associated with minimally invasive techniques.

Although conventional median sternotomy provides optimal surgical exposure, its disruption of sternal integrity may increase postoperative risks of sternal infection and poor wound healing (8,9). Recently, the RAMT approach has gained attention for its unique anatomical advantages. The aortic root is located superficially behind the right sternum, and through a 5–6 cm incision at the 2^nd or 3^rd intercostal space, combined with thoracoscopic assistance, clear exposure of the ascending aortic root and coronary ostia can be achieved. This approach neither requires rib resection nor damages the right internal mammary artery. By optimizing cannulation strategies, deep hypothermic circulatory arrest can be avoided, significantly reducing the risk of neurological complications (7). This institutional study demonstrates that while right anterior exposure is slightly less optimal than median sternotomy for aortic root visualization, precise coronary button reimplantation and graft anastomosis can still be successfully achieved by adjusting thoracoscopic angles and using long-shafted instruments. Particularly in cases with a high-riding right coronary ostium, the right anterior approach offers an advantageous viewing angle that minimizes excessive traction on the aortic wall. However, this technique requires heightened spatial awareness from the surgeon. Literature reports that thoracoscopic-assisted procedures face the challenge of two-dimensional visualization, and restricted instrument angles may complicate suturing (10). In this study, the RAMT group had a longer aortic cross-clamp time, primarily due to the meticulous adjustments required during valve implantation and coronary anastomosis. Innovative surgical techniques, such as the use of sutureless valves and automated anastomotic devices, could help address the technical challenge of prolonged aortic cross-clamp time, potentially reducing manual suturing steps and operative duration (11). Additionally, advanced anastomotic techniques, like the reinforced “French cuff”, may enhance hemostasis and simplify proximal graft attachment (12). Furthermore, there was no difference in prosthetic valve sizes between the groups, suggesting that minimally invasive techniques did not compromise valve selection due to spatial constraints. This effectively mitigates the risk of patient-prosthesis mismatch, consistent with long-term follow-up results from Staromlynski et al. (13).

The pre-matching data in this study showed that the RAMT group had a higher median age (67.5 years) compared to the FS group (59 years, P=0.004), indicating a clinical preference for minimally invasive techniques in elderly patients or those with significant comorbidities. Preserving sternal integrity through minimally invasive surgery helps reduce the risks of postoperative respiratory insufficiency and sternal infection, particularly benefiting high-risk populations such as those with osteoporosis or diabetes (14). Consistent with these findings, Abijigitova et al. demonstrated that upper hemisternotomy reduced pulmonary complication rates by 32% compared to FS in elderly patients (15). However, advanced age often accompanies aortic wall calcification or atherosclerosis, which increases the risk of vascular injury and embolism during minimally invasive procedures. To address this, our center employs preoperative high-resolution computed tomography angiography (CTA) to precisely assess the extent and distribution of aortic calcification. For patients with severe calcification, we prioritize axillary artery cannulation to avoid plaque dislodgement during femoral artery retrograde perfusion. Additionally, intraoperative transesophageal echocardiography (TEE) is used to provide real-time monitoring of guidewire positioning, ensuring cannulation safety. Although acute cerebral infarction was a rare event in this study, Murzi et al. (16) noted that retrograde femoral artery perfusion during minimally invasive procedures may increase the risk of neurological complications in elderly patients, highlighting the importance of rigorous preoperative assessment and optimized cannulation strategies.

The longer aortic cross-clamp time observed in the RAMT group can be attributed to the constrained exposure and instrument maneuverability inherent to the minimally invasive approach, which initially prolongs certain key procedural steps. However, this did not lead to a longer total operative time. This phenomenon can be explained by two factors: First, the minimally invasive approach eliminates the need for sternotomy and sternal closure, reducing chest opening and closing time by approximately 30–60 minutes, which offsets the time spent on meticulous manipulation. Second, when the RAMT cases were grouped sequentially, the CPBT in the later cases was numerically lower than in the earlier cases, consistent with experience-related reductions reported in studies of minimally invasive cardiac procedures. For example, researchers have observed that surgeons often require 20–30 minimally invasive Bentall procedures to achieve operative times comparable to conventional sternotomy (17). In our series, the observed trend aligns with such a learning curve, though the small sample size limits statistical inference. Further analysis revealed that intraoperative blood loss was lower in the RAMT group than in the FS group (P=0.002). This difference may be related to the more extensive tissue exposure required in the FS approach, which often involves greater dissection of muscle, vasculature, and nerve structures. Nevertheless, the FS technique continues to offer distinct advantages in surgical field visualization. It is noteworthy that the reduction in blood loss in the RAMT group did not result in a lower transfusion requirement in our study, while some studies suggest that minimally invasive surgery may reduce transfusion needs (18).

No cases of all-cause mortality or severe acute myocardial infarction occurred in either group. Notably, the RAMT group had no sternal-related complications, such as osteomyelitis or sternal dehiscence, while the FS group had one case of poor wound healing, underscoring the advantage of preserving sternal integrity—particularly in obese or diabetic patients. A multicenter study by Abud et al. demonstrated that mini-thoracotomy reduces the risk of sternal infection by 47% (19), which is consistent with our findings. One case of acute cerebral infarction occurred in the FS group. Given that Bentall surgery and intraoperative CPB may increase the perioperative risk of stroke, this case highlights the ongoing need for vigilance against neurological complications even with conventional procedures. The isolated case of infective endocarditis in the RAMT group underscores the importance of preventing intraoperative instrument contamination. Although the surgical approach cannot be entirely excluded as a contributing factor, existing literature suggests that infective endocarditis is more strongly associated with preoperative oral infections and prolonged postoperative catheterization (20). To mitigate infection risks, our center has implemented stricter aseptic protocols and mandatory preoperative dental consultations for high-risk patients.

At a median follow-up of 12 months, outcomes were satisfactory in all patients except for one case of infective endocarditis. These findings align with the 1-year follow-up data reported by Shah et al. (18), although long-term risks related to bioprosthetic valve degeneration warrant attention. In our center’s RAMT group, the usage rate of bioprosthetic valves was 68.8% (11/16), comparable to the conventional group (75%, 24/32). Although 10-year follow-up studies suggest that minimally invasive approaches do not increase the risk of bioprosthetic structural valve deterioration, age-related calcium metabolism abnormalities in elderly patients may accelerate valve degeneration (13). Therefore, our center implements strict anticoagulation management for bioprosthetic valve patients (target international normalized ratio, 1.8–2.5), with echocardiographic follow-ups at 1, 3, 6, and 12 months during the first postoperative year, followed by annual evaluations to dynamically assess valve morphology and function.

This study has several limitations. First, as a single-center retrospective analysis, the sample size—particularly in the matched cohort—remains relatively limited despite the use of matching to control for confounding. This results in insufficient statistical power for endpoints with low event rates. Therefore, nonsignificant findings should be interpreted with caution and cannot be taken as evidence of complete equivalence between the two groups. Second, the median follow-up period of 12 months is insufficient to evaluate long-term complications, such as valve deterioration or aortic redilation. Future studies should expand the sample size through multicenter collaboration and extend follow-up to over 5 years to comprehensively assess the long-term benefits of minimally invasive techniques.

Conclusions

In summary, this comparative study of the Bentall procedure performed via RAMT and FS revealed no significant differences between the two groups in terms of 30-day MACCEs, postoperative complication rates, or short-term efficacy, indicating that the RAMT approach provides perioperative safety comparable to that of FS. Although aortic cross-clamp time was significantly longer in the RAMT group, intraoperative blood loss was markedly reduced, and sternal-related complications were avoided, which holds important clinical significance for elderly patients or those with high-risk factors such as osteoporosis. However, this technique requires a higher level of surgical skill and poses a discernible technical challenge to master. The median 12-month follow-up results showed favorable outcomes in all patients except for one case of PVE. This study provides clinical evidence supporting RAMT as a feasible minimally invasive option for the Bentall procedure, although its long-term efficacy requires further validation through larger sample sizes and extended follow-up periods.

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. The study was approved by the Ethics Committee of The Second Affiliated Hospital, Zhejiang University School of Medicine (No. 2025-Lunshenyan-0626). Given the retrospective nature of the study and the strict protection of patient privacy, the Ethics Committee waived the requirement for patient informed consent.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2216/dss

jtd-18-02-91-dss.pdf ^{(24.8KB, pdf)}

DOI: 10.21037/jtd-2025-aw-2216