Surgical treatment experience of seven cases of Berry syndrome

Abstract

Objective

Berry syndrome is a group of rare congenital cardiac malformations including aortopulmonary window (APW), aortic origin of the right pulmonary artery (AORPA), interruption of the aortic arch (IAA), patent ductus arteriosus (PDA) (supplying the descending aorta) and intact ventricular septum. This paper will analyze the clinical data of 7 patients with Berry syndrome who underwent surgical treatment in our institution and discuss the one-stage surgical correction of Berry syndrome in combination with the literature.

Methods

From January 2013 to July 2024, a total of 7 children with Berry syndrome were admitted to the Cardiac Surgery Department of Beijing Children’s Hospital. The median age was 3 months (range, 1–36 months). All patients’ IAA morphology were type A. The APW morphology was type IIA in 2, type IIB in 4, and type III in 1 patient. Three different surgical correction techniques were used to repair the APW and AORPA, including intra-aortic patch in 2, RPA angioplasty with aortic cuff in 2, RPA detachment and reimplant in 3 patients.

Results

Among the 7 patients, one died in the early postoperative period, (1/7, 14.3%). The remaining 6 surviving patients, mechanical ventilation was lasted for 51 to 166 h postoperatively, with an average of (113.3 ± 50.8) hours; the CCU stay was 6 to 23 days, with an average of (11.8 ± 6.5) days. Two cases (2/7, 28.6%) of patients adopted the strategy of delayed sternal closure. The 6 surviving children were followed up for a period ranging from 3 to 132 months, with a median follow-up duration of 36 months. During the follow-up, 2 patients underwent a second operations (2/6, 33.3%). The remaining 4 patients showed no obvious RPA stenosis, descending aorta (DAO) stenosis, aortic valve stenosis or aortic valve regurgitation (AR) during the follow-up period. In the latest follow-up, the average velocity of the RPA of the 4 patients was 1.68 ± 0.36 m/s, and the average pressure gradient was 11.9 ± 4.8 mmHg; the average velocity of the DAO was 2.1 ± 1.7 m/s, and the average pressure gradient was 17.9 ± 2.6 mmHg. All the AR were less than mild.

Conclusion

Most children can achieve one-stage surgical correction. For children with APW type IIA, the intra-aortic patch method can be attempted, but its therapeutic effect still requires medium to long-term follow-up. The surgical approach of RPA detachment and reimplant can be applied to all types of patients with Berry syndrome, and the medium to long-term follow-up result is favorable. For the treatment of IAA, it is recommended that end-to-side anastomosis be performed between the DAO and the aortic arch, and the anterior wall be augmented by using bovine pericardial tissue patches. For the residual obstruction at the postoperative anastomosis site, balloon dilation angioplasty can be considered. Compression of the left main bronchus can be supported by intratracheal stents.

Perspective Statement

In the English literature accessed thus far, there are less than 50 cases associated with the surgical treatment of Berry syndrome. In this work, we analyzed the clinical data of 7 patients from January 2013 to July 2024 with Berry syndrome who underwent surgical treatment in our institution and showed one-stage surgical correction can achieved acceptable outcomes.

Graphical Abstract

Background

Berry syndrome is a group of rare congenital cardiac malformations. It was initially described by Berry and his colleagues in 1982, and the first operation was performed [1]. The main pathologies of this malformation include (Fig. 2, A): Aortopulmonary window (APW); aortic origin of the right pulmonary artery (AORPA); Interruption of the aortic arch (IAA); Patent ductus arteriosus (PDA) (supplying the descending aorta); Intact ventricular septum. Its incidence merely constitutes 0.046% of all patients with congenital heart diseases(CHD). This syndrome represents a complex and rare malformation. In the English literature accessed thus far, aside from the group of 16 patients with Berry syndrome reported by Renjie Hu et al. [2], the majority of the remaining literature comprises case reports, and there are less than 50 cases associated with the surgical treatment of Berry syndrome [2, 3]. This paper will analyze the clinical data of 7 patients with Berry syndrome who underwent surgical treatment in our institution from January 2013 to July 2024 and discuss the one-stage surgical correction of Berry syndrome in combination with the literature.

Patients and methods

From January 2013 to July 2024, a total of 7 children with Berry syndrome were admitted to the Cardiac Surgery Department of Beijing Children’s Hospital, among whom 4 were male and 3 were female. The age ranged from 1 to 36 months, with a median age of 3 months. The weight varied from 4 to 13.9 kg, with a median weight of 5 kg. Before the operation, 4 cases were provided with Nasal Continuous Positive Airway Pressure (NCPAP)-assisted breathing due to wheezing and breathing difficulties. None of the children in this group received mechanical ventilation before the operation. 3 cases were administered alprostadil intravenous infusion before the operation to maintain the patency of the PDA. All patients underwent echocardiography and cardiac CT examinations before the operation for definite diagnosis.

Regarding the anatomical classification of IAA, this study also employed the classic classification approach reported by Celoria and Patton [4]. That is, when IAA is located at the distal end of the left subclavian artery, it is defined as type A; when IAA is located between the left common carotid artery and the left subclavian artery, it is type B; when IAA is located between the brachiocephalic trunk and the left common carotid artery, it is type C. For the classification of APW, it was primarily based on the Mori classification and the subsequent classification by Berry on this foundation [1, 5]. Type I indicates that the APW is situated at the proximal end of the ascending aorta (AAO), adjacent to the aortic valve; Type II indicates that the APW is located at the distal end of the AAO; Type III indicates that the APW is large and encompasses the entire AAO. According to the anatomical relationship between right pulmonary artery (RPA) and APW, Type II is further divided into Type IIA and Type IIB. Type IIA indicates that the RPA straddles the APW and still has a continuation with main pulmonary artery (MPA) (Fig. 1, A), while Type IIB indicates that the RPA has no continuation with MPA (Fig. 1, B).

Fig. 1

Preoperative CT showing different Type II of the APW in Berry syndrome. (A), Type IIA, RPA straddles the APW and still has a continuation with MPA; (B), Type IIB, the RPA has no continuation with MPA. Notes: APW, aortopulmonary window; RPA, right pulmonary artery; MPA, main pulmonary artery; AAO, ascending aorta

Operative technique

All pediatric patients underwent one-stage surgical correction procedures. All surgical operations for the patients were performed under cardiopulmonary bypass (CPB). Among this cohort of patients, 3 patients utilized the deep hypothermic circulatory arrest (DHCA) technique, and 4 patients adopted the deep hypothermic low-flow selective cerebral perfusion (DHLF) technique. In our center, the temperature control for both DHCA and DHLF were below 18 °C. The blood flow control for DHLF was 25 ml/kg/min.

All patients were placed in the supine position and underwent a median sternotomy incision. The AAO and its branches, the left and right pulmonary arteries (LPA and RPA), the PDA, and the descending aorta (DAO) were thoroughly dissected. A Y-shaped connecting tube was used to link two arterial cannulas, which were respectively inserted into the arterial perfusion tubes via the distal AAO and MPA. A single right atrial venous cannula or superior and inferior vena cava cannula was employed to establish CPB. The LPA and RPA were occluded by snares. Subsequently, the arterial perfusion tube within the MPA was anterogradely guided into the DAO through the PDA for lower extremity perfusion. After cooling to the target temperature, the aorta was cross-clamped (ACC) and cold crystalloid cardioplegia was infused into the aortic root.

During the intraoperative incision of the anterior wall of the APW, the APW and the opening of the RPA were exposed. According to the different approaches of reconstructing the RPA, they were classified into three groups: Group 1, Group 2, and Group 3. In Group 1, there were 2 cases. The internal tunnel was formed by applying intra-aortic patch, through which the blood flow of the MPA was directly drained into the RPA via the APW (Fig. 2, A, B). Both of the 2 patients utilized Gore-Tex patches. In Group 2, 2 cases were encompassed. The AAO was transected above and below the level of the RPA, leaving its posterior aortic cuff to constitute the posterior wall of the RPA. The anterior wall of the RPA of one patient was augmented by using a bovine pericardial tissue patch, while in the other case, a Gore-Tex patch was utilized. An end-to-end anastomosis was performed on the AAO (Fig. 2, C, D). After the completion of the aforementioned operations, both Group 1 and 2 underwent IAA corrective surgery under DHLF or DHCA. Firstly, the PDA tissue was completely resected, the DAO was fully mobilized, and the DAO was lifted and an end-to-side anastomosis was carried out between the end of the DAO and the incision at the lower edge of the aortic arch. If necessary, a patch was employed to augmented the anterior wall of the anastomosis (Fig. 2, B, D). In Group 1, the anastomoses of 2 patients were directly end-to-side anastomosis. Among the 2 patients in Group 2, the Gore-Tex patch was employed to enlarge and reduce the tension on the anterior wall of the anastomosis for one patient, while the other was directly end-to-side anastomosis. In Group 3, 3 patients were included. The RPA was dissected from the right side of the AAO, and the incision of the RPA was sutured transversely for the reconstruction of the posterior wall of the aorta. The APW was completely incised near the end of the MPA (Fig. 2, E). Proceeding along the incision of the RPA, a continuous anastomosis was conducted to reconstruct the proximal end of the left wall of the AAO. The PDA was transected. The DAO was adequately mobilized. The DAO was elevated, and an end-to-side anastomosis was carried out between the end of the AAO and the incision at the lower edge of the aortic arch. The anterior wall of the anastomosis was enlarged and tension-reduced by using a bovine pericardial tissue patch. The posterior wall of the RPA was directly anastomosed to the MPA, and the anterior wall of the RPA and the MPA was repaired with a bovine pericardial tissue patch (Fig. 2, F).

The general demographic data mentioned above are summarized in Table 1.

Table 1 Summary of demographic characteristics

Fig. 2

Anatomical characteristics of Berry syndrome and different methods of one-stage surgical correction of Berry syndrome. (A), Berry syndrome is a combination of AORPA, APW, PDA, and IAA; (B), apply an intra-aortic patch to guide the blood flow from MPA to RPA via the APW; (C), AAO was transected above and below the level of the RPA, leaving its posterior aortic cuff to constitute the posterior wall of the RPA; (D), the anterior wall of the RPA was augmented by using a patch, an end-to-end anastomosis was performed on the AAO; (E), RPA was transected from the right side of the AAO, APW was completely incised near the end of the MPA; (F), the aortic side of the APW was closed transversely, the posterior wall of the RPA was directly anastomosed to the MPA, and the anterior wall of the RPA and the MPA was repaired with a patch; (B, D,F), an end-to-side anastomosis was carried out between the end of the AAO and the incision at the lower edge of the aortic arch, the anterior wall of the anastomosis was augmented by a patch

Follow up

All patients were obliged to return to the outpatient clinic of cardiac surgery in our hospital at 1 month, 3 months, and 6 months after discharge, and subsequently every 12–18 months for reexamination. Physical examination, echocardiography, and electrocardiogram were utilized to evaluate cardiac function and residual anatomical problems. Cardiac CT examination was supplemented when necessary.

Statistical analysis

SPSS 22.0 (SPSS, IL, USA) was used to analyze all data. Descriptive statistics are reported as means ± SD. Categorical variables are expressed as frequency and percentages. A Kruskal-Wallis test was used to compare the differences among the 3 groups. P values of < 0.05 were considered statistically significant.

Results

The CPB time for the 7 surgical children varied from 120 to 250 min, with an average of (171.8 ± 45.8) minutes; the ACC (aortic cross-clamping) time ranged between 74 and 92 min, with an average of (85.0 ± 7.2) minutes; the DHLF persisted from 31 to 49 min, with an average of (36.8 ± 8.3) minutes; the DHCA lasted from 19 to 43 min, with an average of (34.0 ± 13.1) minutes. There was no statistically significant difference in the CPB and ACC times among the three surgical approaches (P = 0.812, P = 0.055).

Early postoperative outcomes

Among the 7 patients, one died in the early postoperative period, (1/7, 14.3%) (Patient 1). This patient was male, 2 months old, and weighed 5 kg. Cardiac malformations were identified during pregnancy. The patient presented with repeated pneumonia and breathing difficulties after birth. Preoperatively, NCPAP-assisted respiration was actively administered. The preoperative echocardiography and cardiac CT examination of this patient clearly diagnosed: IAA (Type A), APW, AORPA, PDA, and severe pulmonary hypertension. One-stage surgical correction of cardiac malformations was performed under DHCA. Unfortunately, during the operation, the child had difficulty detaching from the CPB and was converted to left heart assistance. The patient died of circulatory failure 24 h after the surgery.

Among the remaining 6 surviving patients, mechanical ventilation was lasted for 51 to 166 h postoperatively, with an average of (113.3 ± 50.8) hours; the CCU stay was 6 to 23 days, with an average of (11.8 ± 6.5) days. Two cases (2/7, 28.6%) of patients adopted the strategy of delayed sternal closure. Apart from the aforementioned patient 1 who died of circulatory failure 24 h after the operation, for the other patient with delayed sternal closure (patient 5), bedside sternal closure was carried out after achieving hemodynamic stability on the third day after the operation. All patients were routinely given sildenafil orally to lower pulmonary artery pressure postoperatively, at a dose of (1 mg/kg, Q8H).

Patient 4, a 3-year-old girl, developed a pulmonary hypertension crisis 4 days postoperatively. The manifestations were sudden hemodynamic instability, presenting as increased heart rate, low blood pressure, low oxygen saturation, and high central venous pressure (CVP). We actively administered sedation, re-performed tracheal intubation, and added treprostinil for reducing pulmonary artery pressure at a dose of 25 ng/kg/min for continuous pump maintenance.

Patient 3 was a 4-month-old girl. She experienced difficulty in being detached from the mechanical ventilator after the operation, and the left lung was atelectatic. Subsequently, through fiberoptic bronchoscopy exploration, it was found that the left main bronchus was significantly compressed. Concurrently, a metal endobronchial stent was placed within the left main bronchus (Fig. 3, A, D). Subsequently, the child was successfully detached from the mechanical ventilator and transferred to the general ward. One month after the operation, a re-examination of fiberoptic bronchoscopy revealed that the endothelization of the endobronchial stent was satisfactory, and there was no granulation hyperplasia (Fig. 3, B, E). Four years subsequent to the surgery, a re-examination of fiberoptic bronchoscopy was conducted. It was found that the endothelialization at the metal endobronchial stent was complete, and the lumen of the left main bronchus was unobstructed (Fig. 3, C, F). One month after the operation, CT re-examination indicated that the DAO after surgical reconstruction anteriorly compresses the left main bronchus, and the morphology of the aortic arch was Roman arch appearance. (Fig. 4, B, C).

The early postoperative outcomes mentioned above are summarized in Table 2.

Fig. 3

(A-C), presents the opening conditions of the left main bronchus following the tracheal stent implantation immediately, 1 month, and 4 years post-implantation; (D-F), exhibit the extent of stent endothelialization and the intraluminal conditions of the left main bronchus subsequent to stent placement immediately, 1 month, and 4 years post-implantation

Table 2 Summary of early postoperative outcomes

Medium and long-term follow-up

Six surviving children were followed up for a period ranging from 3 to 132 months, with a median follow-up duration of 36 months. During the follow-up, two patients underwent a second operations (2/6, 33.3%). The medium and long-term follow-up and reoperation circumstances subsequent to the surgery are summarized in Table 3.

Table 3 Summary of medium and long-term follow-up and reoperations

Patient 3 manifested bicuspid aortic valve deformity, mild aortic stenosis, and moderate aortic valve regurgitation (AR) before the surgery. At the 1-year follow-up after the operation, moderate AR was detected. During the 4-year follow-up visit, echocardiography indicated mild aortic stenosis and moderate AR. At the 7-year outpatient echocardiography and cardiac CT showed moderate aortic stenosis, severe AR, and significant dilation of the AAO were suggested (Fig. 4, D). Subsequently, this patient underwent the Ozaki operation in another hospital and is still under follow-up at present.

Fig. 4

(A-C), CT re-examination indicated the DAO after surgical reconstruction anteriorly compresses the left main bronchus; (C-D), the varying degrees of dilation of AAO 1 month and 7 years postoperatively. Notes: DAO, descending aorta; AAO, ascending aorta; black arrow, the left main bronchus oppressed by the DAO; white arrow, the metallic stent within the left main bronchus; *, the dilation AAO

Patient 2 was followed up in the outpatient department 2 years after the surgery. It was demonstrated that there was mild stenosis of the RPA and moderate stenosis of the DAO. The flow velocity of the RPA was 2.54 m/s, with a pressure gradient of 26 mmHg. The flow velocity of the DAO was 4 m/s, with a pressure gradient of 64 mmHg. Subsequently, the patient was rehospitalized for percutaneous interventional balloon dilation to alleviate the residual obstruction at the postoperative anastomosis site of DAO (Fig. 5).

Fig. 5

(A), Evident stenosis is present at the anastomotic site of the DAO; (B), Balloon dilation angioplasty of the anastomotic site of the DAO

The remaining 4 patients showed no obvious RPA stenosis, DAO stenosis, aortic valve stenosis or AR during the follow-up period. In the latest follow-up, the average velocity of the RPA of the 4 patients was 1.68 ± 0.36 m/s, and the average pressure gradient was 11.9 ± 4.8 mmHg; the average velocity of the DAO was 2.1 ± 1.7 m/s, and the average pressure gradient was 17.9 ± 2.6 mmHg. All the AR were less than mild.

Discussion

The pathological alterations of Berry syndrome predominantly involve the APW and the AORPA, thereby giving rise to a significant left-to-right shunt at the AAO level. In infants with this syndrome, continuous pulmonary hypertension emerges in the early postnatal stage. The pathological anatomy of IAA leads to the blood supply to the lower extremities of children being completely dependent on the PDA. Due to the specific and fatal pathophysiological changes, such patients require radical surgery promptly upon diagnosis to prevent the persistent pulmonary hypertension and the lower extremity blood supply relying on the PDA [6].

In developed countries and regions, due to the wide application of fetal echocardiography, the majority of children can be diagnosed during the maternal pregnancy period and receive timely treatment after birth [7,8,9]. Nevertheless, in China, numerous regions still witness a relatively low level of prenatal examination and an insufficient understanding of CHD. A considerable number of children with CHD fail to obtain timely diagnosis and treatment. The median age of the children in this group at the time of surgery was 3 months. The PDA of children with Berry syndrome is usually rather large, a minority still need preoperative administration of alprostadil to maintain the patency of the PDA. However, the majority of children do not require emergency surgery due to PDA closure. It is still recommended to conduct surgical correction under relatively comprehensive preoperative examinations and preparations.

The principle of surgical treatment is to eliminate the APW, reestablish the physiological blood flow from the MPA to the RPA, transected the PDA and restore the physiological blood flow from the aortic arch to the DAO. The key point of the surgical discussion resides in the reconstruction of the RPA [2, 10]. Currently, three surgical approaches are employed in clinical practice: (1) Utilizing an intra-aortic patch to form a tunnel for directly channeling the venous blood from the MPA into the RPA through the APW; (2) The AAO was transected above and below the level of the RPA, leaving its posterior aortic cuff to constitute the posterior wall of the RPA. The anterior wall of the RPA is directly sutured or augmented with a patch; (3) Disconnecting the RPA and retransplanting it onto the MPA. Each of the three surgical modalities has its own advantages and disadvantages.

The advantages of the surgical approach of forming a tunnel with intra-aortic patch are as follows: (1) The surgical procedure is relatively simple. Two problems, namely the APW and the AORPA, can be simultaneously addressed by one patch; (2) The continuity between the RPA and the MPA is maintained, eliminating the necessity for transplantation of the right pulmonary artery; (3) The incision of the AAO is reduced, thereby minimizing the probability of postoperative bleeding. The disadvantages are: (1) The patch is located within the AAO. If the patch is overly large, it is prone to cause stenosis of the AAO. If the patch is too small, it is prone to result in stenosis of the RPA; (2) The pressure within the AAO after the surgery is higher than that in the RPA. The patch may protrude towards the opening of the RPA, leading to RPA obstruction; (3) The patch has no growth potential. Residual obstruction may occur in the medium to long term after the operation, and it is rather challenging to conduct secondary intervention through percutaneous interventional methods. As the size, nature, and location of the patch will have an impact on the blood flow of the AAO and the RPA, the following issues need to be taken into account when performing the patch: Firstly, the distance between the opening of the RPA and the APW; Secondly, whether the blood flow between the RPA and the MPA is continuous; Thirdly, the size of the APW. According to the classification of APW, for children of type IIA, the RPA straddles the APW and is in continuity with the MPA, and the opening position of the RPA is close to the APW. Therefore, the tunnel intra-aortic patch method is applicable. For children of type IIB, since the RPA is located at the right posterior of the AAO and there is a lack of continuity between the RPA and MPA, and the RPA opening position is far from the APW. In this situation, if the internal tunnel patch is used, the patch would be overly large, thereby influencing the blood flow of the AAO. Mannelli and colleagues described several cases of intra-aortic patch repair in type IIB patients with no complications [6]. The experience reported by Shi et al. showed post-operative RPA stenosis, unfortunately after applying this method [10].

In this cohort of pediatric patients, two cases were managed with the intra-aortic patch approach. These two cases were the first and second instances of Berry syndrome undergoing surgical treatment in our center. In the initial years, owing to the immaturity of surgical techniques and inadequate knowledge of the disease, both patients manifested diverse problems postoperatively. Case 1 died of circulatory failure 24 h after the operation due to difficulty in weaning from the CPB. In Case 2, during the one-year postoperative follow-up, the echocardiography of the heart revealed that the velocity of the RPA escalated to 2.54 m/s, with a pressure gradient of 26 mmHg, and the velocity of the descending aorta increased to 4 m/s, with a pressure gradient of 64 mmHg. A percutaneous balloon dilation procedure was carried out to alleviate the stenosis of the anastomosis of the DAO. At present, the explicit surgical indications and contraindications for the intra-aortic patch approach remain undefined in clinical practice. Further follow-up is essential to observe the medium and long-term efficacy. The two patients in our center who underwent this surgical modality at an earlier stage, and currently, this surgical approach is no longer utilized in our center.

The second surgical approach for reconstructing the RPA entails incising the orifice of the RPA along with the posterior wall of the aorta to the APW, using it as the posterior wall of the RPA [11]. The anterior wall is either directly sutured or augmented by a patch. In our group of cases, two patients employed this surgical approach. The merits of this surgical modality are that the continuity between the RPA and the MPA is preserved. The drawbacks are as follows: (1) Transection of the AAO is necessary, resulting in relatively significant surgical trauma and a high risk of postoperative bleeding. Particularly, active bleeding from the posterior wall of the aorta might lead to irreversible and catastrophic consequences. (2) This surgical approach causes a shortening of the AAO in length and a reduction in the height of the aortic arch, thereby exerting compression on the RPA and the left main bronchus. If an artificial conduit is employed to elongate the AAO, a new anastomosis will be added, increasing the risk of bleeding. This surgical method is highly challenging and demands high proficiency from the surgeon. During the operation, the length of the AAO, the diameter of the RPA, and their relationship need to be evaluated simultaneously, otherwise, it is prone to induce multiple postoperative complications such as postoperative bleeding, residual obstruction, and airway compression. Postoperative RPA stenosis caused by the compression from the AAO was identified in one patient reported by Renjie Hu et al. [2]. This patient previously received RPA arterioplasty with aortic cuff and a direct end-to-end aortic anastomose. Lecompte maneuver was performed on postoperative day 7 to release the compression of the RPA. Shi et al. reported that RPA is anastomosed to the MPA at the anterior of the aorta with patch augmentation in a similar fashion to the LeCompte maneuver [10].

The third surgical approach for the reconstruction of the RPA is to directly excise the RPA from the posterior wall of the AAO and reimplant it into the MPA, the posterior wall of the RPA was directly anastomosed to the MPA, and the anterior wall of the RPA and the MPA was repaired with a patch [12]. Currently, this surgical modality has also been the principal one utilized by our center in recent years. The advantage lies in the fact that there is no necessity to consider the position of the orifice of the RPA and the size of the APW. It can be applicable to all children with Berry syndrome. However, Duyen et al. report a case of Berry syndrome, in whom myocardial ischaemia developed following direct implantation of the RPA to the MPA, which was resolved using an interposition tube graft [13]. Our center has made certain amendments to the classic surgical method. After excising the RPA from the right side of the AAO, the incision of the RPA is immediately sutured transversely to reconstruct the posterior wall of the AAO. The APW is completely incised near the end of the MPA, and the incision resembles a trapdoor (Fig. 2, E). The anterior wall tissue of the APW is employed to reconstruct the proximal end of the left wall of the AAO. The benefit of this is that there is no demand for additional patches for the AAO, which shortens the operation duration and reduces bleeding. The posterior wall of the RPA is directly sutured to MPA, and bovine pericardial tissue patches are used to repair the anterior walls of the RPA and MPA to alleviate the tension of the anastomosis. Currently, there are 3 patients in this center who have undergone this surgical procedure. The postoperative recoveries of all of them are satisfactory. Neither AR nor postoperative complications such as stenosis of the RPA and trachea stenosis have emerged.

Here we share some experiences of our center regarding the surgical management of different APWs. For simple APW, the experience of our center is to incise the anterior wall of the APW near the MPA. After completely dissected the APW, the lateral wall of the AAO is directly sutured, and a patch is used to repair the MPA. We do not recommend inserting a patch to separate the APW using the sandwich method, as the resulting morphology might be poor, and the different pressures on both sides of the patch could lead to localized stenosis in the contralateral vascular lumen. For APW with AORPA, the intra-aortic patch can be attempted for type IIA, but it may still cause localized stenosis of the pulmonary artery. For type IIB, the third method mentioned in the paper is adopted to correct APW and AORPA simultaneously.

For the correction of IAA, the surgical approach is similar to that for isolated IAA. Based on our experience, it is recommended to perform end-to-side anastomosis between the DAO and the aortic arch. The anterior wall is augmented by utilizing a bovine pericardial tissue patch to alleviate the tension at the anastomotic site. Literature reports suggest that the residual obstruction of the aortic arch can lead to proximal dilation of the aortic arch, thereby compressing the RPA and the left main bronchus [14]. For the residual obstruction of the aortic arch, percutaneous interventional balloon dilation can be contemplated [14]. One patient in our group had difficulty in weaning off the ventilator after the operation, along with atelectasis of the left lung. Subsequently, through fiberoptic bronchoscopy exploration, it was revealed that the left main bronchus of the child was significantly compressed after the operation. A metallic endobronchial stent was concurrently placed in the left main bronchus. Subsequently, the child successfully weaned off the ventilator and was transferred to the general ward. One year after the operation, a CT re-examination indicated that the morphology of the aortic arch presented a Roman arch appearance, and the DAO significantly compressed the left main bronchus anteriorly. At the 7-year follow-up after the operation, the child had a considerable amount of AR. On the one hand, this patient had bicuspid aortic valve deformity and moderate aortic regurgitation before the operation. On the other hand, it might be attributed to the change in the morphology of the aortic arch caused by the operation, and the poor forward blood flow further exacerbated the AR of this patient. Subsequently, this patient underwent the Ozaki operation in another hospital and is currently still under follow-up.

Limitations

This study possesses numerous limitations. To commence with, it is a retrospective study, and all samples originate from a single center. Despite Berry syndrome being a rare congenital heart disease, the 7 patients incorporated in this group cover a considerable time span. Due to the certain discrepancies in treatment concepts, surgical technical proficiency, cardiopulmonary bypass levels, and postoperative monitoring levels during different time periods, the prognosis of patients diagnosed and treated in distinct time intervals may vary as a result of the aforementioned differences. As a consequence of the uneven distribution of medical resources, the screening levels for prenatal and postnatal congenital heart diseases in many areas of China are relatively backward. The patients in this group are all relatively aged, with no neonatal cases. The average age of the patients in this group is 3 months. Regarding the stenosis of the RPA and the DAO and the AR in the long term after different surgical approaches, further follow-up is still necessary.

Conclusions

The perioperative mortality rate of Berry syndrome is relatively high. Once diagnosed, radical surgical treatment should be carried out as soon as possible. Most children can achieve one-stage surgical correction. For children with APW type IIA, the intra-aortic patch method can be attempted, but its therapeutic effect still requires medium to long-term follow-up. The surgical approach of RPA detachment and reimplant can be applied to all types of patients with Berry syndrome, and the medium to long-term follow-up result is favorable. For the treatment of IAA, it is recommended that end-to-side anastomosis be performed between the DAO and the aortic arch, and the anterior wall be augmented by using bovine pericardial tissue patches. For the residual obstruction at the postoperative anastomosis site, balloon dilation angioplasty can be considered. Compression of the left main bronchus can be supported by intratracheal stents.