Abstract
Background
Limited-access surgical repair remains important for adults with large atrial septal defects (ASDs) when transcatheter closure is unsuitable. This research aimed to compare early outcomes of inferior mini-sternotomy and right axillary vertical mini-incision for large ASD repair under cardiopulmonary bypass (CPB) without aortic cross-clamping.
Methods
This retrospective study included 79 adults from January 2020 to January 2024. Based on surgical approaches, patients were divided into Group A (n = 40, inferior mini-sternotomy) and Group B (n = 39, right axillary vertical mini-incision). Operative variables, postoperative recovery, pain scores, complications, costs, and 3-month follow-up findings were compared.
Results
All patients underwent successful repair without perioperative death, conversion, reoperation, major residual shunt, stroke, or low-output syndrome. Operative time, CPB duration, transfusion volume, duration of tracheal intubation, ICU stay, postoperative hospital stay, hospitalization cost, and myocardial injury markers were comparable between groups. Group B had higher pain scores on postoperative days 2 and 3 (P < 0.05). Pulmonary infection, pneumothorax, pleural effusion, and phrenic nerve palsy occurred only in Group B, whereas superficial wound infection, sternal malunion, and hypertrophic scar or keloid occurred only in Group A; most complication differences were not statistically significant. During 3-month follow-up, no death, reoperation, major residual shunt, cerebrovascular event, or severe rhythm-related event was observed.
Conclusions
Both approaches enabled safe early large ASD repair on-pump CPB without aortic cross-clamping. Right axillary vertical mini-incision offered better scar concealment but greater early pain and pulmonary-management demands.
Keywords: Atrial septal defects, Cardiopulmonary bypass without aortic cross-clamping, Surgical wound, Mini-sternotomy, Axillary vertical mini-incision
Introduction
An atrial septal defect (ASD) is one of the most prevalent congenital heart malformations, making up approximately 10 ~ 15% of congenital heart defects [1]. Transcatheter closure has been widely used in many heart centres, frequently replacing conventional surgical repair [2]. However, to ensure patient safety, surgical repair remains essential for large or anatomically complex ASDs unsuitable for device closure [3]. Although conventional median sternotomy provides reliable exposure, it is associated with greater sternal trauma, a visible anterior scar, postoperative pain, and wound- or sternum-related morbidity [4, 5]. Limited-access approaches have therefore been increasingly used in ASD repair. Mini-sternotomy preserves a familiar midline operative route and facilitates central cannulation and cardiac exposure [6, 7], whereas the right axillary vertical mini-incision avoids anterior sternal division and offers a concealed lateral scar [8, 9]. These two approaches differ in anatomical access, their perioperative profiles may differ in pain, pulmonary events, wound healing, and cosmetic burden. In addition, selected cardiac repairs can be performed on a beating heart under cardiopulmonary bypass (CPB) without aortic cross-clamping [10], but comparative evidence on different limited-access routes under this strategy remains limited. Therefore, this study compared the early outcomes of inferior mini-sternotomy and right axillary vertical mini-incision for large ASD repair using a standardized beating-heart CPB strategy without aortic cross-clamping.
Materials and methods
Study population
This retrospective single-center study reviewed the medical records of adult patients who underwent ASD surgical repair at the Department of Cardiovascular Surgery, the First People’s Hospital of Guangyuan between January 2020 and January 2024. A total of 79 patients were included and divided according to surgical approach into Group A (inferior mini-sternotomy), and Group B (right axillary vertical mini-incision). All procedures were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the 1975 Declaration of Helsinki, as revised in 2000. Ethical approval was obtained from The First People’s GuangYuan Hospital (No.2022−080). Written informed consent was obtained from the patients and their families.
Inclusion and exclusion criteria
The inclusion criteria were as follows: (1) age ≥ 18 years; (2) diagnosis of a large ASD confirmed by transthoracic echocardiography and/or transesophageal echocardiography; (3) ASD transverse diameter ≥ 20 mm; (4) unsuitable anatomy for transcatheter closure after preoperative evaluation; (5) complete perioperative clinical data.
The exclusion criteria were as follows: (1) primum ASD; (2) ASD combined with complex intracardiac malformations requiring additional complex cardiac procedures, such as ventricular septal defect, tetralogy of Fallot, or severe valve disease requiring surgical correction; (3) severe pulmonary hypertension or Eisenmenger syndrome; (4) severe extracardiac organ disease requiring simultaneous treatment; (5) previous cardiac surgery; (6) incomplete clinical records; (7) refusal to participate by the patient or family. Patients with sinus venosus ASD, simple anomalous pulmonary venous return, or simple concomitant tricuspid valvuloplasty that could be repaired during the same operation were not excluded.
Preoperative evaluation
All patients underwent routine assessment including physical examination, electrocardiography, chest radiography, blood and biochemical tests, and echocardiography. Transthoracic echocardiography (TTE) and/or transesophageal echocardiography (TEE) was used to evaluate ASD size, anatomical subtype, rim adequacy, shunt status, Qp/Qs ratio, right-sided chamber enlargement, tricuspid regurgitation, pulmonary artery pressure, echocardiographically estimated pulmonary vascular resistance, and suitability for transcatheter closure. Surgical repair was selected when transcatheter closure was considered unsuitable because of deficient rims, excessively large defects, multiple defects, sinus venosus ASD, or anomalous pulmonary venous return.
Selection of surgical approach
Patients with right thoracic disease, recent or active pulmonary infection after clinical stabilization, pleural adhesion, or previous right thoracic surgery were preferentially assigned to the mini-sternotomy group. For patients without these contraindications, the approach was chosen through shared decision-making between the surgical team and the patient, based on sex, body habitus, thoracic anatomy, cosmetic preference, and the surgeon’s judgment of operative exposure.
Operative technique
All procedures were performed by the same surgical team under general anesthesia and CPB. Intraoperative TEE was routinely performed in all patients after induction of anesthesia to reconfirm ASD morphology, venous drainage, valve function, ventricular function, and associated intracardiac abnormalities. A beating-heart strategy without aortic cross-clamping was used in all patients. After systemic heparinization, CPB was established by ascending aortic cannulation and bicaval venous cannulation under direct vision. The superior and inferior venae cavae were snared after initiation of CPB to reduce venous return to the right atrium. A restrictive transfusion strategy was applied in both groups, with a hemoglobin threshold of 70 g/L and a target range of 70–90 g/L. Concomitant tricuspid valvuloplasty, when required, was performed according to intraoperative findings.
Group A: inferior mini-sternotomy (Fig. 1, and Fig. 2) Patients were placed in the supine position, and standard single-lumen endotracheal intubation was used. A 7 ~ 11 cm lower midline skin incision was made over the lower sternum. An inferior inverted L-shaped partial sternotomy was performed by dividing the sternum from the xiphoid process upward to the right second intercostal space below the sternal angle, followed by right-sided transverse extension. The pericardium was opened longitudinally and suspended to expose the ascending aorta, superior vena cava, inferior vena cava, and right atrium.
Fig. 1.
Inferior mini-sternotomy in group A
Fig. 2.
Intraoperative repair of a large ASD through an inferior mini-sternotomy. Cardiopulmonary bypass was established with ascending aortic and bicaval venous cannulation. The ASD was repaired on the beating heart without aortic cross-clamping using a Dacron/polyester patch. AO, ascending aorta; SVC, superior vena cava; IVC, inferior vena cava; ASD, atrial septal defect
After CPB was established, the right atrium was opened on the beating heart without aortic cross-clamping. The ASD was inspected directly, including its size, location, rim quality, and relationship with the venae cavae and pulmonary venous drainage. The defect was repaired using a Dacron or autologous pericardial patch according to defect size, rim adequacy, and tissue condition; Dacron patches were preferred for large defects requiring stable structural support. The patch was sutured continuously with polypropylene sutures. After completion of the repair, de-airing was performed, the right atriotomy was closed, CPB was discontinued after hemodynamic stabilization, and a mediastinal or pericardial drainage tube was placed before wound closure.
Group B: right axillary vertical mini–incision (Fig. 3) Patients were placed in the left lateral position with the right hemithorax elevated approximately 30 degrees, and the right arm was fixed above the head to expose the axillary region. After induction of general anesthesia, double-lumen endotracheal intubation was used to allow left one-lung ventilation and right lung collapse during right thoracic entry and exposure. A 7 ~ 11 cm vertical incision was made along the right infra-axillary region. The skin, subcutaneous tissue, and muscle layers were separated along the muscle fibers as much as possible to avoid unnecessary muscle transection. The thoracic cavity was entered through the fourth intercostal space. During entry and exposure, particular attention was paid to protecting the breast tissue, intercostal vessels and nerves, right lung, and phrenic nerve. The pericardium was opened and suspended to bring the right atrium closer to the operative field.
Fig. 3.
Right vertical infra-axillary incision was performed in group B
CPB was then established under direct vision. After snaring of the venae cavae, the right atrium was opened on the beating heart without aortic cross-clamping. The ASD was inspected and repaired with a Dacron or autologous pericardial patch using continuous polypropylene sutures. After repair, careful de-airing was performed, the right atrium was closed, and CPB was discontinued after stable circulation was achieved. A right thoracic drainage tube was placed through a separate small incision, usually at the seventh intercostal space along the mid-axillary line, before closure of the thoracic and skin incisions.
Because all procedures were performed on a beating heart without aortic cross-clamping, strict measures were used to prevent air embolism. The operative field was flooded with carbon dioxide, and the patient was placed in a Trendelenburg position when appropriate. Continuous suction was used to control the intracardiac blood level. During ASD repair, the blood level in the atrium was maintained at approximately the level of the atrial septum or slightly lower, but not below the mitral valve plane. Before right atrial closure and separation from CPB, meticulous de-airing was performed through the operative field and suction lines. Intraoperative TEE was used to assess intracardiac air, patch position, valve function, ventricular function, and residual shunting before termination of CPB.
Data collection
Perioperative data were retrospectively extracted from the electronic medical record system by two independent investigators and cross-checked for accuracy. The collected variables included demographic characteristics, ASD anatomical subtype and diameter, reasons for surgical referral, echocardiographic and hemodynamic parameters, operative time, CPB duration, blood transfusion volume, incision length, duration of tracheal intubation, ICU stay, postoperative drainage volume, chest tube removal time, postoperative hospital stay, hospitalization cost, myocardial injury markers, postoperative rhythm outcomes, VAS pain scores, and postoperative complications.
The primary outcomes were surgical success and postoperative complications. Surgical success was defined as complete ASD repair without large residual shunt, low-output cardiac syndrome, cerebrovascular accident, severe rhythm-related complication, or reoperation. Large residual shunt was defined as residual shunt > 3 mm on postoperative echocardiography. Postoperative complications were classified as cardiac, pulmonary, neurological, and wound-related events. Cardiac complications included new-onset atrial fibrillation (AF), persistence of preoperative AF, transient atrioventricular block, pericardial effusion, low-output cardiac syndrome, severe ventricular arrhythmia, and large residual shunt. Pulmonary complications included pulmonary infection, atelectasis, pneumothorax, pleural effusion, and phrenic nerve palsy. Pulmonary infection specifically referred to clinically diagnosed pneumonia, rather than pleural infection or empyema, and was defined as the presence of at least two of the following criteria: new or progressive pulmonary infiltrates on chest radiography, fever > 38.0 °C, purulent sputum, or leukocyte count > 12 × 10⁹/L or < 4 × 10⁹/L. Wound-related complications included superficial wound infection, sternal malunion, hypertrophic scar or keloid, and subcutaneous emphysema.
The secondary outcomes included operative and early postoperative recovery indicators, myocardial injury markers, and VAS pain scores on postoperative days 1, 2, and 3.
All patients were followed at 1 month and 3 months after surgery. Follow-up included physical examination, electrocardiography, and TTE performed by the same sonographer. Follow-up was completed in all 79 patients, with no loss to follow-up.
Statistical analyses
Continuous variables were expressed as means±standard deviations. The quantitative data were analyzed using t-tests or rank-sum tests. The count data were analyzed using the Fisher’s exact test. VAS scores were compared between groups at postoperative days 1, 2, and 3. All statistical analyses were performed using IBM SPSS Statistics, Version 27.0 (IBM Corp., Armonk, NY, USA). A p-value of < 0.05 was deemed to be statistically significant.
Results
Baseline characteristics
A total of 79 adult patients with ASD were included. Secundum ASD was the predominant subtype, while a small proportion had sinus venosus ASD. All patients had been evaluated for transcatheter closure before surgery. The most common reason for surgical referral was rim deficiency, followed by excessive defect size, multiple defects, and anomalous pulmonary venous return. The distribution of referral reasons was comparable between Group A (n = 40) and Group B (n = 39). The two groups showed comparable baseline demographic, anatomical, and hemodynamic characteristics, including age, sex, body weight, cardiothoracic ratio, ASD diameter and subtype, preoperative AF, tricuspid regurgitation, reasons for surgical referral, left atrial and right atrial dimensions, right ventricular diameter, TAPSE, Qp/Qs ratio, pulmonary artery pressure, and pulmonary vascular resistance (all P > 0.05; Table 1). The results indicated well-matched baseline atrial remodeling, right-sided chamber enlargement, right ventricular systolic function, and pulmonary hemodynamics.
Table 1.
Comparison of preoperative data between the two groups of patients
| Item | Group A (n = 40) | Group B (n = 39) | P-value | |
|---|---|---|---|---|
| Age (y) | 39.90 ± 11.03 | 37.00 ± 15.07 | 0.339 | |
| Sex (M/F) | 15/25 | 17/22 | 0.936 | |
| Body weight (kg) | 57.18 ± 8.90 | 59.93 ± 8.64 | 0.174 | |
| ASD type | Secundum | 36 | 34 | 0.735 |
| Sinus venosus (SVC type) | 3 | 4 | ||
| Sinus venosus (IVC type) | 1 | 1 | ||
| Size of ASD (mm) | Transverse diameter | 26.66 ± 7.13 | 28.82 ± 9.04 | 0.249 |
| Vertical diameter | 21.05 ± 5.83 | 21.16 ± 6.66 | 0.698 | |
| Evaluated for transcatheter closure | 40/40 | 39/39 | — | |
| Reason for surgical referral | Deficient rim(s) | 28 | 26 | 0.836 |
| Defect too large (> 36 mm) | 8 | 9 | ||
| Multiple defects | 3 | 3 | ||
| Anomalous pulmonary venous return | 1 | 1 | ||
| Left atrial diameter (mm) | 36.42 ± 5.18 | 37.15 ± 5.76 | 0.556 | |
| Right atrial dimension (mm) | Transverse | 45.83 ± 6.24 | 47.12 ± 7.01 | 0.393 |
| Longitudinal | 53.26 ± 6.87 | 54.08 ± 7.43 | 0.615 | |
| Right ventricular diameter (mm) | 32.45 ± 5.12 | 33.21 ± 5.58 | 0.527 | |
| TAPSE (mm) | 19.82 ± 3.14 | 19.35 ± 3.47 | 0.531 | |
| PVR (Wood units) | 2.34 ± 0.86 | 2.51 ± 0.93 | 0.405 | |
| Pulmonary artery pressure (mmHg) | 35.69 ± 11.16 | 37.46 ± 14.00 | 0.543 | |
| CTR | 0.54 ± 0.04 | 0.56 ± 0.06 | 0.222 | |
| Qp/Qs | 2.19 ± 0.36 | 2.17 ± 0.35 | 0.808 | |
| With AF | 9/40 | 8/39 | 0.812 | |
| With TR | Mild | 15/40 | 14/39 | 0.725 |
| Moderate/severe | 7/40 | 9/39 | 0.453 |
ASD, atrial septal defect; M/F, male/female; CTR, cardiothoracic ratio; AF, atrial fibrillation; TR, tricuspid regurgitation; LAD, left atrial diameter; RA, right atrium; RVD, right ventricular diameter; TAPSE, tricuspid annular plane systolic excursion; PVR, pulmonary vascular resistance; SVC, superior vena cava; IVC, inferior vena cava
Operative and early postoperative outcomes
All patients underwent successful ASD repair under CPB without aortic cross-clamping. Intraoperative TEE was performed in all patients and confirmed satisfactory patch position without major residual shunt before separation from CPB. There was no perioperative death, conversion to full median sternotomy, or reoperation. There were also no significant differences in operative time, CPB duration, duration of tracheal intubation, ICU stay, postoperative drainage volume, chest tube removal time, blood transfusion volume, incision length, postoperative hospital stay, or hospitalization cost (all P > 0.05; Table 2). Peak CK-MB and peak cTnI levels were comparable between groups, suggesting no obvious between-group difference in perioperative myocardial injury under the beating-heart strategy.
Table 2.
Comparison of perioperative and postoperative data between the two patient groups
| Item | Group A (n = 40) | Group B (n = 39) | P-value |
|---|---|---|---|
| Duration of surgery (min) | 191.44 ± 28.08 | 192.50 ± 24.31 | 0.859 |
| CPB duration (min) | 45.82 ± 12.68 | 48.95 ± 11.94 | 0.268 |
| Duration of tracheal intubation (h) | 8.33 ± 1.83 | 8.29 ± 1.37 | 0.901 |
| ICU stay (h) | 18.42 ± 2.68 | 18.39 ± 2.87 | 0.955 |
| Postoperative drainage (mL) | 206.10 ± 69.51 | 194.38 ± 61.33 | 0.434 |
| Chest tube removal time (h) | 38.65 ± 8.72 | 36.82 ± 7.93 | 0.338 |
| Blood transfusion (mL) | 141.28 ± 28.35 | 151.25 ± 30.21 | 0.841 |
| Incision length (cm) | 8.09 ± 1.02 | 7.98 ± 1.01 | 0.617 |
| Postoperative hospital stay (days) | 6.87 ± 1.30 | 6.50 ± 1.12 | 0.182 |
| Cost of hospitalization (×10⁴ RMB) | 4.18 ± 0.51 | 4.05 ± 0.35 | 0.173 |
| Myocardial injury markers | |||
| Peak CK-MB (U/L) | 18.36 ± 6.42 | 19.12 ± 7.05 | 0.614 |
| Peak cTnI (ng/mL) | 0.87 ± 0.34 | 0.92 ± 0.38 | 0.538 |
| Postoperative rhythm outcomes | |||
| New-onset AF | 2/40 | 1/39 | 1.000 |
| Preoperative AF persisted postoperatively | 5/9 | 4/8 | 1.000 |
| Concomitant surgical ablation performed | 0/9 | 0/8 | — |
| VAS score | |||
| Postoperative day 1 | 4.03 ± 0.78 | 4.35 ± 0.92 | 0.098 |
| Postoperative day 2 | 3.42 ± 0.65 | 5.04 ± 0.87 | < 0.001 |
| Postoperative day 3 | 3.18 ± 0.58 | 4.71 ± 0.82 | < 0.001 |
RMB, Chinese Yuan; CPB, cardiopulmonary bypass; VAS, Visual Analogue Scale; CK-MB, creatine kinase-MB; cTnI, cardiac troponin I; AF, atrial fibrillation
The AF burden (Group A, 9/40; Group B, 8/39), right atrial and right ventricular enlargement, preserved right ventricular systolic function (TAPSE > 17 mm), and mildly elevated PVR in both groups suggest chronic atrial volume overload and atrial remodeling (Table 1). Further differentiation revealed that among the 17 patients with preoperative AF, postoperative persistence of AF was observed in 5 of 9 patients in Group A and 4 of 8 patients in Group B (P = 1.000), and new-onset AF occurred in 2 patients in Group A and 1 patient in Group B (P = 1.000) (Table 2). No concomitant surgical ablation was performed in either group. VAS pain scores were similar on postoperative day 1, but were significantly higher in Group B on postoperative days 2 and 3 (both P < 0.001), indicating greater early postoperative pain after the right axillary vertical mini-incision approach.
Postoperative complications and follow-up
All 79 patients completed 1- and 3-month follow-up. Neither group experienced 30-day mortality, low-output cardiac syndrome, severe ventricular arrhythmia, cerebrovascular accident, large residual shunt, or reoperation (Table 3). The incidences of new-onset AF, transient atrioventricular block, pericardial effusion, atelectasis, pneumothorax, pleural effusion, phrenic nerve palsy, and wound-related complications were not significantly different between groups (all P > 0.05). Pulmonary infection occurred only in Group B (4/39 vs. 0/40), but the between-group difference did not reach statistical significance (P = 0.054). Sternal malunion, superficial wound infection, and hypertrophic scar/keloid were observed only in Group A, while pneumothorax, pleural effusion, and phrenic nerve palsy occurred only in Group B. These events were managed conservatively, and no patient required additional surgical intervention.
Table 3.
Comparison of postoperative complications between the two patient groups
| Item | Group A (n = 40) | Group B (n = 39) | P-value |
|---|---|---|---|
| Cardiac complications | |||
| Large residual shunt | 0 | 0 | — |
| Low-output cardiac syndrome | 0 | 0 | — |
| Severe arrhythmia (VT/VF) | 0 | 0 | — |
| New-onset AF | 2 | 1 | 1.000 |
| Transient AV block | 1 | 0 | 1.000 |
| Pericardial effusion | 3 | 2 | 1.000 |
| Pulmonary complications | |||
| Pulmonary infection | 0 | 4 | 0.054 |
| Pneumothorax | 0 | 1 | 0.494 |
| Atelectasis | 1 | 3 | 0.358 |
| Pleural effusion | 0 | 1 | 0.494 |
| Neurological complications | |||
| Cerebrovascular accident | 0 | 0 | — |
| Phrenic nerve palsy | 0 | 1 | 0.494 |
| Wound-related complications | |||
| Subcutaneous emphysema | 0 | 0 | — |
| Sternal malunion | 1 | 0 | 1.000 |
| Wound infection (superficial) | 1 | 0 | 1.000 |
| Hypertrophic scar/keloid | 1 | 0 | 1.000 |
| Reoperation for any cause | 0 | 0 | — |
| 30-day mortality | 0 | 0 | — |
Fisher’s exact test was used for all categorical variables. P < 0.05 was defined as statistically significant. Large residual shunt: residual shunt > 3 mm on postoperative echocardiography
Pulmonary infection was defined as the presence of at least two of the following criteria: (1) new or progressive pulmonary infiltrates on chest radiography; (2) fever > 38.0 °C; (3) purulent sputum; (4) leukocyte count > 12 × 10⁹/L or < 4 × 10⁹/L. This definition was applied in accordance with the Centers for Disease Control and Prevention (CDC) criteria for clinically defined pneumonia.
Discussion
Transcatheter closure is preferred for anatomically suitable secundum ASDs; however, surgical repair remains essential for patients with deficient rims, excessively large or multiple defects, sinus venosus ASD, or anomalous pulmonary venous return, in whom device closure carries risks of deployment failure, residual shunt, and other complications [2, 11, 12]. In the present study, all 79 patients underwent preoperative assessment for transcatheter closure and were ultimately referred for surgery because of unfavorable anatomical conditions. Under a standardized strategy of beating-heart repair on cardiopulmonary bypass without aortic cross-clamping, both limited-access approaches achieved successful repair, with no perioperative death, conversion to full median sternotomy, reoperation, major residual shunt, cerebrovascular event, or low-output cardiac syndrome. These findings suggest that this technical framework has favorable early safety in adult patients with large ASD.
Conventional open intracardiac surgery relies on aortic cross-clamping and cardioplegic arrest to obtain a motionless operative field, but this process introduces global myocardial ischemia-reperfusion and increases the burden of myocardial protection [13]. ASD repair is confined to the atrial septum through a right atrial approach and does not involve complex left-sided structures, thus avoiding additional myocardial injury and circulatory stress, which is particularly relevant in adult large ASD patients with chronic volume overload, right-heart enlargement, and atrial remodeling [2, 12]. Avoiding aortic cross-clamping may reduce ischemia-reperfusion injury and decrease the technical complexity of aortic manipulation through a limited incision [14–16]. Previous studies have shown that beating-heart ASD repair can achieve favorable early safety and repair efficacy [17, 18]. In the study, comparable CK-MB and cTnI levels, together with the absence of low-output cardiac syndrome, severe ventricular arrhythmia, or stroke, further supported the early feasibility of the beating-heart strategy without aortic cross-clamping. The major concern of no-cross-clamp repair is air embolism [17, 19]. Therefore, this study used carbon dioxide flooding, Trendelenburg positioning when necessary, continuous negative-pressure suction, controlled atrial blood level, and meticulous de-airing to reduce this risk.
Limited-access approaches differ in cardiac exposure, potentially affecting cosmetic and clinical outcomes [20]. In the study, with comparable baseline characteristics and a unified beating-heart strategy, the two approaches demonstrated similar operative efficiency, early recovery, and severe complication rates, supporting their early safety. Group A preserved a familiar midline operative axis, allowing direct exposure of the ascending aorta, venae cavae, and right atrium, and facilitating central cannulation and right atrial manipulation [21]. However, this approach still involves partial sternotomy and an anterior midline chest incision, which was consistent with the sternal malunion, superficial wound infection, and hypertrophic scarring observed in a small number of patients [22, 23]. By contrast, Group B avoided sternal division and placed the scar in a concealed lateral thoracic location [8, 9, 24], but required intercostal and pleural access. Intercostal access, retraction, instrument pressure, and thoracic drainage may irritate or injure intercostal nerves and contribute to postoperative pain [25, 26], which explains the higher VAS scores on postoperative days 2 and 3 in this group. Pain-related impairment of coughing and sputum clearance may increase atelectasis and pneumonia risk [27]. In addition, although double-lumen intubation facilitates lung isolation and exposure, one-lung ventilation, pleural entry, lung re-expansion, and thoracic drainage may increase postoperative pulmonary-management demands [28, 29]. These mechanisms may partly explain why pulmonary infection, pneumothorax, pleural effusion, and phrenic nerve palsy were observed only in Group B. This highlights the need for stronger analgesia, lung protection, and respiratory rehabilitation when using the right axillary approach. Additionally, despite the relatively young age in both groups, approximately half of the patients with preoperative AF remained in AF after surgery. This suggests that ASD repair reduces shunting and volume overload, but established atrial enlargement and arrhythmogenic substrate caused by chronic volume overload may persist after anatomical correction [2, 30]. Therefore, preoperative rhythm assessment and long-term rhythm monitoring should be considered in adult ASD patients with AF.
Clinically, surgical access should be individualized according to ASD anatomy, exposure requirements, thoracic condition, pulmonary risk, and cosmetic preference. Inferior mini-sternotomy may be more appropriate for patients with right thoracic disease, recent pulmonary infection or increased pulmonary risk, prior right chest surgery, or large defects requiring direct assessment of caval or pulmonary venous relationships. The key technical requirements include safe exposure for ascending aortic and bicaval cannulation, reliable caval snaring, right atrial retraction, and patch suturing, with adequate pericardial suspension and stable sternal retraction to reduce bleeding and incision-related complications [21, 31]. By contrast, the right axillary vertical mini-incision may be considered for patients with favorable thoracic anatomy, low pulmonary risk, and strong preference for scar concealment. The technical priorities include accurate intercostal space selection, dissection along muscle fibers, protection of the breast tissue and intercostal vessels and nerves, adequate pericardial suspension to shorten the operative distance, and placement of thoracic drainage through a separate small incision to reduce intercostal nerve irritation [24, 31]. Regardless of the selected approach, limited-access surgery should not focus solely on incision concealment, but also on safe tissue exposure, stable CPB, controlled intracardiac blood flow, reliable patch fixation, repeated de-airing, and absence of major residual shunt [32]. Postoperative care should include enhanced analgesia, early lung expansion, active sputum clearance, and rhythm monitoring, particularly after the right axillary approach.
This study had several limitations. First, it was a retrospective single-center study with a small sample size, and the surgical approach was not randomly assigned; therefore, selection bias and limited statistical power for uncommon complications cannot be excluded. Second, although baseline anatomical and hemodynamic variables were balanced, unmeasured factors such as thoracic morphology, pulmonary reserve, and surgeon preference may still have influenced access selection. Third, follow-up was limited to 3 months, preventing assessment of long-term right-heart remodeling, atrial fibrillation evolution, residual shunting, chronic pain, scar satisfaction, and quality of life. Additionally, although postoperative rhythm outcomes were reported, patients with preoperative atrial fibrillation did not undergo concomitant surgical ablation, and standardized long-term rhythm monitoring was not performed; therefore, the true postoperative atrial fibrillation burden and rhythm evolution could not be fully assessed. Finally, the absence of a conventional full-sternotomy or arrested-heart control group precludes direct comparison with traditional repair. Future prospective multicenter studies with longer follow-up are needed to validate these findings.
Conclusions
In selected adult patients with large ASD unsuitable for transcatheter closure, both inferior mini-sternotomy and right axillary vertical mini-incision enabled safe early repair under CPB without aortic cross-clamping. The two approaches showed comparable repair efficacy, but different access-related burdens: inferior mini-sternotomy offered more direct exposure with fewer pleural complications, whereas the right axillary approach provided better cosmetic concealment but greater early pain and pulmonary-management demands. Our experience suggests that an individualized approach selection can be made based on factors such as the anatomical structure of the defect, exposure requirements, thoracic conditions, pulmonary risks, and patient preferences.
Author contributions
CX: Conceptualization, Methodology, Software, Investigation, Formal Analysis, Writing - Original Draft; YH: Data Collection and Curation, Writing - Original Draft; XW: Figures; JC: Tables, Resources, Supervision; YZ: Statistical Analyses, Validation; CM: Visualization, Writing - Review & Editing; CM: Conceptualization, Funding Acquisition, Resources, Supervision, Writing - Review & Editing. All authors read and approved the final manuscript.
Funding
The authors have no funding sources to declare.
Data availability
Complete research data can be provided. If needed, please send an email to corresponding author to request it.
Declarations
Ethics approval and consent to participate
The research reported had adhered to the relevant ethical guidelines.
Consent for publication
Written informed consent was obtained from the patient for publication of this study and accompanying images.
Competing interests
The authors declare no competing interests.
Conflicts of interest/Competing interests
The authors have no conflicts of interest to disclose.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Amabile A, DegifeE, Krane M, LaLonde MR, Gruber PJ, Geirsson A. Robotic, totally endoscopic atrial septal defect repair.[J]. Multimedia manual of cardiothoracic surgery: MMCTS, 2021, 2021. [DOI] [PubMed]
- 2.Brida M, Chessa M, Celermajer D, Li W, Geva T, Khairy P, Griselli M, Baumgartner H, Gatzoulis MA. Atrial septal defect in adulthood: a new paradigm for congenital heart disease.[J]. Eur Heart J. 2022;43(28):2660–71. [DOI] [PubMed] [Google Scholar]
- 3.Faccini A, Butera G. Atrial septal defect (ASD) device trans-catheter closure: limitations.[J]. J Thorac disease. 2018;10(Suppl 24):S2923–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sabzi F, Faraji R, Kazeminasab M. Minimal Invasive Technique in Atrial Septal Defect Surgery.[J]. Cardiol Res. 2018;9(2):90–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Maeßen T, Korir N, Van de Velde M, Kennes J, Pogatzki-Zahn E, Joshi GP. Pain management after cardiac surgery via median sternotomy: A systematic review with procedure-specific postoperative pain management (PROSPECT) recommendations.[J]. Eur J Anaesthesiol. 2023;40(10):758–68. [DOI] [PubMed] [Google Scholar]
- 6.El-Andari R, Watkins AR, Fialka NM, AlAklabi AM, Kang JJH, Bozso SJ, Jonker D, Freed DH, AlAklabi M, Bouchard D, Chu MWA, Moon MC, Nagendran J. A Systematic Review of Minimally Invasive Approaches to Surgical Atrial Septal Defect Repair.[J]. Heart Lung Circ. 2025;34(8):764–76. [DOI] [PubMed]
- 7.Konstantinov IE, Buratto E. Atrial Septal Defect Closure via Ministernotomy in Children.[J]. Heart Lung Circ. 2021;30(9):e98–100. [DOI] [PubMed] [Google Scholar]
- 8.Yaliniz H, Topcuoglu MS, Gocen U, Atalay A, Keklik V, Basturk Y, Gunes Y, Turktan M, Salih OK. Comparison between minimal right vertical infra-axillary thoracotomy and standard median sternotomy for repair of atrial septal defects.[J]. Asian J Surg. 2015;38(4):199–204. [DOI] [PubMed] [Google Scholar]
- 9.Yang X, Hu Y, Dong J, Huang P, Luo J, Yang G, Deng X. Rightvertical axillary incision for atrial septal defect: a propensity score matched study.[J]. J Cardiothorac Surg. 2022;17(1):256. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Güllü AÜ, Şenay Ş, Ersin E, Demirhisar Ö, Whitham T, Koçyiğit M, Alhan C. Robotic-assisted cardiac surgery without aortic cross-clamping: A safe alternative approach.[J]. J Card Surg. 2021;36(1):165–8. [DOI] [PubMed] [Google Scholar]
- 11.Baumgartner H, De Backer J, Babu-Narayan SV, Budts W, Chessa M, Diller G, Lung B, Kluin J, Lang IM, Meijboom F, Moons P, Mulder BJM, Oechslin E, Roos-Hesselink JW, Schwerzmann M, Sondergaard L, Zeppenfeld K. 2020 ESC Guidelines for the management of adult congenital heart disease.[J]. Eur Heart J. 2021;42(6):563–645. [DOI] [PubMed] [Google Scholar]
- 12.Alkashkari W, Albugami S, Hijazi ZM. Current practice in atrial septal defect occlusion in children and adults.[J]. Expert Rev Cardiovasc Ther. 2020;18(6):315–29. [DOI] [PubMed] [Google Scholar]
- 13.Sabe SA, Harris DD, Broadwin M, Sellke FW. Cardioprotection in cardiovascular surgery.[J]. Basic Res Cardiol. 2024;119(4):545–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gaudino M, Benesch C, Bakaeen F, DeAnda A, Fremes SE, Glance L, Messé SR, Pandey A, Rong LQ. Considerations for Reduction of Risk of Perioperative Stroke in Adult Patients Undergoing Cardiac and Thoracic Aortic Operations: A Scientific Statement From the American Heart Association.[J]. Circulation. 2020;142(14):e193–209. [DOI] [PubMed] [Google Scholar]
- 15.Chiari P, Fellahi J. Myocardial protection in cardiac surgery: a comprehensive review of current therapies and future cardioprotective strategies.[J]. Front Med. 2024;11:1424188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Yun T, Kim H, Sohn B, Chang HW, Lim C, Park K. Robot-Assisted Repair of Atrial Septal Defect: A Comparison of Beating and Non-Beating Heart Surgery.[J]. J chest Surg. 2022;55(1):55–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Tang Y, Wu Y, Zhu J, Liu X, Zhou J, Huang H, Li M, Dai Y, Han X. Total endoscopic repair of atrial septal defect under on-pump beating heart.[J]. J Thorac disease. 2018;10(12):6557–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wang X, Sun H, Ma B, Liu K, Ma Z. Thoracoscopic closure of atrial septal defect in perfused beating hearts[J]. Surg Endosc. 2025;39(1):341–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Dang Q, Le N, Nguyen C, Tran D, Nguyen D, Nguyen T, Ngo T. Totally Endoscopic Cardiac Surgery for Atrial Septal Defect Repair on Beating Heart Without Robotic Assistance in 25 Patients.[J]. Innovations (Philadelphia, Pa.), 2017, 12(6): 446–452. [DOI] [PMC free article] [PubMed]
- 20.Luo Z, Chen Q, Yu L, Chen L, Huang Z. Comparative Study between Surgical Repair of Atrial Septal Defect via Median Sternotomy, Right Submammary Thoracotomy, and Right Vertical Infra-Axillary Thoracotomy.[J]. Brazilian J Cardiovasc Surg. 2020;35(3):285–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mylonas KS, Ziogas IA, Evangeliou A, Hemmati P, Schizas D, Sfyridis PG, Economopoulos KP, Bakoyiannis C, Kapelouzou A, Tzifa A, Avgerinos DV. Minimally Invasive Surgery vs Device Closure for Atrial Septal Defects: A Systematic Review and Meta-analysis.[J]. Pediatr Cardiol. 2020;41(5):853–61. [DOI] [PubMed] [Google Scholar]
- 22.Huang L, Chen D, Chen L, Xu Q, Zheng Z, Dai X. The use of the Scar Cosmesis Assessment and rating scale to evaluate the cosmetic outcomes of totally thoracoscopic cardiac surgery.[J]. J Cardiothorac Surg. 2020;15(1):250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Jina H, Simcock J. Median sternotomy scar assessment.[J]. N Z Med J. 2011;124(1346):57–62. [PubMed] [Google Scholar]
- 24.An K, Li S, Yan J, Wang X, Hua Z. Minimal Right Vertical Infra-axillary Incision for Repair of Congenital Heart Defects.[J]. Ann Thorac Surg. 2022;113(3):896–902. [DOI] [PubMed]
- 25.Dodge-Khatami J, Noor R, Riggs KW, Dodge-Khatami A. Mini right axillary thoracotomy for congenital heart defect repair can become a safe surgical routine.[J]. Cardiol Young. 2023;33(1):38–41. [DOI] [PubMed] [Google Scholar]
- 26.Miyazaki T, Doi R, Matsumoto K. Post-thoracotomy pain syndrome in the era of minimally invasive thoracic surgery.[J]. J Thorac disease. 2024;16(5):3422–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Feray S, Lemoine A, Aveline C, Quesnel C. Pain management after thoracic surgery or chest trauma.[J]. Minerva Anestesiol. 2023;89(11):1022–33. [DOI] [PubMed] [Google Scholar]
- 28.Patel M, Wilson A, Ong C. Double-lumen tubes and bronchial blockers.[J]. BJA Educ. 2023;23(11):416–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Sha Y, Xu R, Shao S, Yang J, Tang B, Liang Q, Wang Z. Tailored single-lung ventilation approaches and postoperative pulmonary outcomes in thoracic surgery.[J]. J Thorac disease. 2025;17(7):5371–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Minaev AV, Mironenko MY, Dontsova VI, Pirushkina YD, Berdibekov BS, Voynov AS, Sarkisyan JA, Golukhova EZ. Left Atrial Structural and Functional Changes in Adults with Congenital Septal Defects and Paroxysmal Atrial Fibrillation.[J]. Journal of clinical medicine, 2024, 13(19). [DOI] [PMC free article] [PubMed]
- 31.Hong Z, Chen Q, Lin Z, Zhang G, Chen L, Zhang Q, Cao H. Surgical repair via submammary thoracotomy, right axillary thoracotomy and median sternotomy for ventricular septal defects.[J]. J Cardiothorac Surg. 2018;13(1):47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hardman G, Zacharias J. Minimal-Access Atrial Septal Defect (ASD) Closure.[J]. J Cardiovasc Dev disease, 2023;10(5):205. [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Complete research data can be provided. If needed, please send an email to corresponding author to request it.