Outcomes of Arterial Switch Operation in Children with Transposition of Great Arteries Older than Two Weeks

INTRODUCTION

In transposition of the great arteries (TGA), the aorta arises from the right ventricle and the pulmonary artery (PA) from the left ventricle (LV).^[1] If untreated, 90% of babies with Dextro TGA (d-TGA) do not survive the 1^st year of life. The Arterial Switch Operation (ASO) was first performed by Dr Adib Dominos Jatene in 1975. Since then, mortality and morbidity have decreased considerably due to improvements in equipment, techniques (e.g., the LeCompte maneuver), and imaging. At present, the ASO described by Jatene, using the Lecompte maneuver,is considered the gold standard for the treatment of TGA. The transfer of the coronary arteries during ASO is a critical factor in determining the procedure’s success.^[2] Children undergoing ASO sometimes require extracorporeal support in the immediate postoperative period for the failing LV. This extra therapy adds to the morbidity and mortality of children. The ASO, although complex, has good results when performed within the first 2 weeks of life.

Our study aimed to investigate the outcomes of the ASO in children older than 2 weeks with TGA, performed by a single surgeon.

The primary objective was to determine 30 days inhospital mortality after surgery. The secondary objectives were to evaluate the duration of intensive care unit (ICU) and hospital stays, postoperative complications, and their correlation with outcomes.

MATERIAL AND METHODS

After institutional ethics committee approval, the medical records of children older than 2 weeks who underwent ASO for TGA by a single surgeon between April 2021 and April 2023 were reviewed retrospectively. All 31 children received the same standardized anesthetic management and conduct of cardiopulmonary bypass (CPB). Surgery was performed under moderate hypothermia (rectal temperature 28°C). After weaning from CPB, controlled ultrafiltration was performed to remove inflammatory mediators (cytokines) from the blood and to reduce fluid accumulation in the extracellular space. Despite these measures, if the child had severe myocardial edema with or without hemodynamic compromise, the sternal closure was delayed until the patient was stabilized. Children were placed on extracorporeal membrane oxygenation (ECMO) either in the operating room (as an aid to come off CPB) or in the ICU (for low cardiac output syndrome). The low cardiac output syndrome in these children has been defined in our earlier publication.^[3]All the children were shifted to the ICU in the postoperative period.

A tracheostomy was performed in children who were hemodynamically stable, afebrile and on mechanical ventilation for more than 7 days with no expectation of extubation in the next 72 h.

RESULTS

Of the 31 TGA children who underwent ASO, 23 were male, and eight were female. The mean age was 4.62 ± 7.85 months, and the mean weight was 4.13 ± 1.86 kg. Seventeen children (54.8%) underwent balloon atrial septostomy (BAS), and one patient underwent BAS + Coarctation of the aorta repair before ASO. Out of 31 children, 19 had a “prepared” LV, 13 had a ventricular septal defect (VSD), and 13 children required ECMO support in the postoperative period. The mean aortic cross-clamp time and CPB time were 103.13 ± 15.65 min and 173.77 ± 30.92 min, respectively.

In an earlier publication from our institute, a “prepared LV” has been defined as where the LV is D-shaped, and the intact ventricular septum (IVS) moves with LV, whereas a banana-shaped LV, where IVS moves with RV, is said to be “regressed.”^[6]Surgery in children who had a regressed LV was at risk of developing left ventricular failure and, therefore, was planned with the support of ECMO in the postoperative period.

The 30-day mortality rate was 32.2% (10/31). For analysis, children were divided into two groups: Survivors and non-survivors at 30-day post-surgery. There were no significant differences between the two groups in demographic characteristics, preoperative procedures, and intraoperative characteristics, i.e., CPB and aortic cross-clamp time (P > 0.05). Furthermore, there were no significant differences in the duration of mechanical ventilation, perioperative ECMO requirement, re-exploration, need for tracheostomy, and incidence of VAP [Table 1].

AKI occurred in 90% of non-survivors vs. 23.8% of survivors (P = 0.001). Non-survivors had a significantly higher (P < 0.001) incidence of postoperative complications, septic shock, intrapulmonary hemorrhage, or refractory cardiogenic shock [Table 1].

The total duration of hospital stay was significantly shorter in the non-survivor group (12.50 ± 5.54 vs. 40.43 ± 25.47 days). Among the survivors, 16 children were alive at 1 year after surgery, and five were lost to follow-up.

Several postoperative variables were significantly associated with 30-day mortality. Non-survivors had significantly higher ISs (27.00 ± 9.49 vs. 18.24 ± 6.80; P = 0.007), higher total drain output (36.34 ± 13.12 vs. 24.32 ± 11.41 mL/kg; P = 0.026), and significantly shorter ICU stay (5.56 ± 5.57 vs. 18.76 ± 15.99 days; P = 0.012) compared to survivors, reflecting early deterioration leading to death [Table 1].

DISCUSSION

In this single-center retrospective study, we evaluated the outcomes of the ASO performed in patients with TGA who were older than 2 weeks. The mean age of the children at the time of surgery was 4.62 months (median - 2 months). In India, children often present late, with further delay in surgery due to various reasons. A lack of resources in developing countries often leads to delayed diagnosis and late presentation.^[7] In our study, 9 children were 2–4 weeks old, and 22 were more than 1 month old at the time of surgery. In almost all cases, children with TGA and an intact ventricular septum undergo BAS to address the immediate issue of insufficient oxygenation. In our study, 18 of 31 children with TGA and an IVS had undergone BAS.

The ASO should ideally be performed in the 1^st week of life to minimize left ventricular deconditioning. In the retrospective cohort study by Anderson et al., 140 patients underwent ASO, with a median age at operation of 5 days (range 1–12 days). It was concluded that delaying the ASO beyond 3 days of age increases major morbidity by 47% and healthcare costs by 8%.^[8] However, all the children in our study were older than 12 days.

The 30-day mortality rate in our children was 32.2%. In our cohort, the age of children at the time of surgery was more than 2 weeks (median age 2 months) and children who underwent BAS were operated on more than 24 h later. The mortality is high when compared to published results of ASO in neonates from high-volume centers (typically <10%).^[9]However, when comparing studies involving late presenters, such as the study by Jonas et al., which included 127 patients who underwent an ASO from 1977 to 2020 at a single center, the outcomes are comparable. Their surgical mortality rate during the entire study period was 24.6%.^[10]

Out of 18 children who underwent BAS and later ASO, 10 children (55%) required ECMO, and of those 10 children, 5 expired. Three (22.3%) out of 13 children with TGA and VSD who required ECMO support died. Overall, 13 of 31 patients (42.6%) required ECMO postoperatively, and the survival rate in this subset was 53.8% (7 of 13). The use of ECMO support in children did not differ significantly between survivors and non-survivors, as the decision to use ECMO was based on the ventricle’s preparedness before surgery and the development of low cardiac output syndrome in the postoperative period. Yang et al. performed a retrospective study of 16 pediatric patients with d-TGA requiring ECMO support from July 2007 to Dec 2019. Two patients were on ECMO support in the preoperative period, whereas 14 patients required ECMO support only postoperatively. Of these 14 children, seven survived to discharge (50% survival).^[11]

No significant association was found between age, weight, body surface area (BSA), and 30-day mortality. There was no effect of age on survival because all our patients underwent surgery well beyond the recommended age of <2 weeks. Furthermore, preoperative BAS in our children did not affect mortality. Hamzah et al. studied 17,392 neonates to evaluate preoperative BAS as a risk factor for mortality in children undergoing ASO. They also did not find any significant difference in mortality between patients who underwent preoperative BAS vs those who did not (6.3% vs 6.7%; P = 0.29).^[12] Similarly, intraoperative variables – including cardiopulmonary bypass time and aortic cross-clamp time – did not differ significantly between survivors and non-survivors, suggesting that surgical and technical factors were consistent across the cohort.

In our study, AKI and other postoperative complications (sepsis, left ventricular failure, etc.) were associated with increased mortality. Postoperative parameters such as IS, AKI, and total drain output also showed strong associations with 30-day mortality.

Higher ISs in the non-survivor group likely reflect more significant myocardial dysfunction postoperatively and are an established predictor of poor outcomes following neonatal and infant cardiac surgery.^[13-15] Prolonged ICU and hospital stay in the survivor group further indicate the complexity of postoperative recovery in these high-risk patients.

Postoperative complications – notably septic shock, intrapulmonary hemorrhage, refractory cardiogenic shock, and AKI were significantly associated with 30-day mortality. Multi-organ dysfunction syndrome (MODS) as a result of long cardiopulmonary bypass and sepsis is a major determinant of mortality after ASO and other complex congenital cardiac surgeries.^[16] In a study done by Ben-Abraham et al. in 722 pediatric patients undergoing congenital cardiac surgery, MODS was the major cause of mortality in the 1^st post-operative week.^[17]

The incidence of AKI in non-survivors was significantly higher than in survivors (P < 0.001). There are multiple reasons for developing AKI in children undergoing ASO – immature kidneys, preoperative medications like diuretics, antibiotics, contrast agents (BAS), dehydration (diarrhea, fever, viral infections), CPB time more than 2 h systemic inflammatory response syndrome (SIRS), vasopressors, low cardiac output state, hypothermia, post-operative bleeding and blood transfusion, infection, liver dysfunction and nephrotoxic medicines in postoperative period. AKI is a well-documented independent risk factor for morbidity and mortality in patients undergoing ASO and may reflect a low cardiac output state, renal hypoperfusion, and systemic inflammation postoperatively - all of which may have contributed to poorer outcomes in our cohort.^[18-20]

CONCLUSION

The mortality in children who are more than 2 weeks old and undergo an ASO is 32.2%. In this cohort, age and preoperative BAS are not associated with 30-day mortality. AKI and other postoperative complications, such as sepsis and left ventricular failure, were associated with increased mortality. Postoperative parameters, such as IS and total drain output, also showed strong associations with 30-day mortality.