Total thoracoscopic repair of ventricular septal defect: A single‐center experience

Abstract Objectives To explore the safety and efficacy of total thoracoscopic repair of ventricular septal defects (VSD). We compared clinical outcomes of VSD via a total thoracoscopic approach with those of mini‐sternotomy. Methods We retrospectively reviewed clinical data from patients with VSD from 2012 to January 2019. According to the surgical pattern, they were divided into two groups: the total thoracoscopic surgery group (36 patients, 27 females, aged 29 ± 9.52 years), and a mini‐sternotomy group (31 patients, 12 females, aged 28 ± 8.67 years). Results There were no deaths in either group. In the thoracoscopic group, cardiopulmonary bypass (CPB) time and aortic cross‐clamping (ACC) time were significantly longer than those of the mini‐sternotomy group (CPB time: 112 ± 23.16 min vs. 78 ± 37.90 min, respectively, p < .001; ACC time: 65 ± 19.94 min vs. 50 ± 24.90 min, respectively, p < .001). postoperative hospital stay time (5.11 ± 2.48 days vs. 5.90 ± 6.27 days, p = .488) and chest drainage (139.86 ± 111.71 ml vs. 196.13 ± 147.34 ml, p = .081) tended to be lower in the thoracoscopy group, although there was no significant difference. No residual shunt or tricuspid regurgitation was found at follow‐up. Conclusions Total thoracoscopic repair is safe and effective in patients with VSD, with or without tricuspid regurgitation.

Nevertheless, in China, due to conservative tendencies and insufficient publicity regarding improved cosmesis, there are worries regarding the costs of increased cardiopulmonary bypass (CPB) duration and possible worse patient outcomes, even though total thoracoscopic repair has shown high rates of successful repair, low morbidity and decreased length of stay in hospital. 5 Therefore, in the present study, we explored the safety and efficacy of total thoracoscopic repair of VSDs, and compared the clinical outcomes of total thoracoscopic with those of mini-sternotomy. The vena cava were isolated with separate tourniquet snares, similar to sternotomy-based surgery. The right common femoral artery was cannulated using a 17-or 19-Fr arterial cannula. The ascending aorta was cross-clamped with a transthoracic aortic cross-clamp and antegrade cardioplegia was delivered into the aortic root, while the body temperature dropped to 32°C. A midbody right atriotomy was made after the superior and inferior vena cava were blocked. If the VSD could be exposed directly, it was closed with a patch of autologous pericardium or a bovine patch. If the VSD was inadequately exposed using the transatrial approach, detachment of the tricuspid valve was performed. The septal tricuspid valve was partially detached by a circumferential parallel incision 2 mm away from the annulus, and the septal leaflet was suspended by 3 or 4 sutures ( Figure 2). After the VSD was continuously sutured with a patch, the septal leaflet was reattached to the annulus with a continuous suture, with the patch sandwiched between the leaflet and the annulus.
Finally, the tricuspid valve coaptation and competence were assessed by injecting the cold saline into the right ventricle. The right thoracic cavity was flooded with CO 2 via the second port throughout F I G U R E 1 Three ports. The first port (2.5-3 cm) was positioned in the fourth intercostal space outside the right midclavicular line. The second port (1-1.5 cm) was placed in the fourth intercostal space anterior axillary line. The third port (1.0-2.0 cm) was located in the fifth intercostal space between the midaxillary line and anterior axillary line. A tissue retractor was insert into the port immediately if each port was made the procedure to avoid gas embolisms. Transesophageal echocardiography was used in each patient immediately after the VSD repair.

| Mini-sternotomy approach
In the mini-sternotomy group, patients underwent a partial midline sternotomy for surgical exposure. 6 Extracorporeal circulation was applied with central arterial and venous cannulation. The myocardial protection and the VSD repair were similar to that performed in the conventional sternotomy procedure. Transesophageal echocardiography was also used in each patient.

| Statistical analysis
The data were prospectively collected in a computerized database, and then analyzed by Statistical Package for Social Sciences, version 22.0 (SPSS Inc.). All continuous variables were expressed as the mean ± SDs if normally distributed, otherwise median and interquartile. And the differences were tested for significance using the Student t test if normally distributed, otherwise Mann-Whitney's U test. The differences in categorical variables were tested for significance with the χ 2 or Fisher exact test, as appropriate, and were presented as proportions unless stated otherwise. The p < .05 was considered to be statistically significant in results.

| RESULTS
During the study period, 67 patients underwent minimally invasive VSD repair at our institution, of whom 36 underwent total thoracoscopic repair and 31 underwent mini-sternotomy repair. There was no significant difference in basic preoperative data between the groups, except for the higher proportion of women in the total thoracoscope group (75% vs. 38.7%; p = .003). Preoperative echocardiography showed 33 perimembranous VSDs and three inlet VSDs in total thoracoscopic group, while four perimembranous VSDs and 27 doubly committed and juxta-arterial VSD. See Table 1 for details.
All patients underwent surgery on an elective basis. The mean CPB and ACC time in the total thoracoscopy group was significantly longer than that of the mini-sternotomy group. Five patients in the total thoracoscopy group underwent tricuspid valvuloplasty with Edward ring followed by VSD repair. They suffered from tricuspid regurgitation due to annular or right ventricular dilatation as a result of left-to-right shunt. In 24 patients, for whom exposure of VSD was difficult due to septal leaflets attached to the rim of the VSD to form an aneurysm or abnormal chordal attached to the septum, underwent tricuspid valve detachment in the total thoracoscopy group.
They were all free from atrioventricular conduction block and tricuspid regurgitation during the follow-up period. See Table 2 for the relevant data.

| 2215
There were no inpatient deaths in either group ( Note: Data presented as mean ± SD or n (%).
Chinese surgeons tried to perform total thoracoscopic procedures without the aid of robots. At present, almost all clinical studies have shown that the mortality and complication rates of total thoracoscopic surgery are not inferior to those of median thoracotomy, and the former affords less bleeding, faster recovery and less trauma.
Our institution's comparative outcomes also support the conclusion that total thoracoscopic VSD repair results in similar excellent results as those of mini-sternotomy. Although the mean CPB and ACC time in the total thoracoscope group were significantly longer than Perhaps traditionalists remain concerned that detachment may increase the incidence of iatrogenic complications such as atrioventricular conduction block and tricuspid valve insufficiency. However, tricuspid valve detachment has been previously shown excellent outcomes. [14][15][16][17] Our results also suggest the detachment was a safe

| LIMITATIONS
This study was subject to the typical limitations of a retrospective case series. There were several limitations, including small sample size and differences in the patient populations. Most patients underwent VSD repair in infancy and early childhood, and we excluded patients with severe pulmonary arteria systolic pressure and other concomitant diseases mentioned in the discussion section. As a result, the sample size was small. In addition, while most preoperative characteristics were well matched, there were significant differences in sex and VSD types among the two groups. More females opted for the total thoracoscopic procedure due to the need for favorable cosmesis. We excluded the subarterial or supracristal VSDs in the thoracoscopy group, which led to the vast majority proportion of this type being in the mini-sternotomy group.

| CONCLUSION
VSD repair can be performed safely and effectively via total thoracoscopy with excellent outcomes similar to those of ministernotomy.