Anesthesia for bilateral pulmonary banding as part of hybrid stage I approach palliating neonates with hypoplastic left heart syndrome

Neonatal management of patients with hypoplastic left heart syndrome and complex remains a challenging task, whereby the “hybrid” palliation is often reserved for high‐risk patients as a “rescue” procedure.


| INTRODUC TI ON
Hypoplastic left heart syndrome (HLHS) and newborns with multiple left heart obstructions, summarized as hypoplastic left heart complex (HLHC), represent a group of rare heart defects with a highly conjoined patient risk profile requiring cardiac surgery during the first days of live. Based on variable hypoplasia of the left ventricle, connected valves, ascending aorta, and aortic arch, the generation of a sufficient perfusion of the systemic circuit is hindered after the physiologic postnatal adaptation process. Surgical treatment for HLHS was expanded from stepwise Norwoodlike procedure and cardiac transplant by the hybrid approach in 1998. 1 The hybrid approach in pediatric cardiovascular surgery is a palliative strategy for HLHS/HLHC and related entities, usually performed in the first, latest two weeks after birth. 2 The combined catheter-based and surgical approach includes an openchest bilateral pulmonary artery banding (bPAB), also as part of cardiac insufficiency therapy, combined with intravenous prostaglandin E1 infusion. Stenting of the ductus arteriosus Botalli (DAB) is performed either via transpulmonary access or using a separate percutaneous approach, both securing systemic perfusion with additional atrial septostomy, if necessary. 3 Despite the fact that hybrid stage I is carried out differently, cardiopulmonary bypass (CPB) and deep hypothermic cardiac arrest (DHCA) can be avoided minimizing surgical trauma during the first days of life and lessening initial operative risks. 4 Nonetheless, a recent Meta-analysis has described inferior early outcomes in patients undergoing hybrid stage I procedure compared to the Norwood procedure, 5 bringing the perioperative management and patient selection into the focus of interest. Cao et al observed a higher risk profile among the patients undergoing the hybrid procedure with an expectable higher incidence of clinically decompensated conditions prior to surgery. 5 It is well known that beside anatomic considerations and comorbidities, the preoperative physiological status has impact on the outcome of patients undergoing stage I Norwood operation or its variants. 6,7 This relationship has not been investigated for patients undergoing a hybrid stage I procedure. The objective of the presented retrospective study was to describe the perioperative anesthetic management and to assess a potential association of the preoperative physiological status (ie, preoperative inotrope requirement) on inhospital mortality, duration of postoperative mechanical ventilation, postoperative time at the intensive care unit, perioperative vasoactive medication requirements, and red blood cell transfusion in patients undergoing bPAB as part of the Giessen stage I hybrid approach (GHA).

| Patient characteristics and perioperative data
We identified 185 neonates (77 female, 108 male) that underwent bPAP as a part of GHA between 1998 and 2015. All patients with HLHS/HLHC who were presented at the GPHC were treated with the hybrid approach. Two small pieces of a PTFE tube graft were used for bPAP. In general, the size of the bands placed around the pulmonary arteries is standardized with 3 mm < 3.0 kg, and 3.5 mm ≥ 3.0 kg, however in individual cases, the surgeon decides on the basis of the intraoperative situs. The clinical course of all patients is shown in Figure 1.
A total of 165 patients (89%) were assigned to Group A and 20 to Group B (11%). The biometric and perioperative data are presented in Table 1. No differences between the groups could be observed with regard to the majority of postnatal risk factors, including the number of patients with a birth weight less than 2.5 kg, prematurity, comorbidities, number and timing of atrial septostomy or atrial stenting, the occurrence of aortic atresia, and surgical procedure time.

| Anesthetic and hemodynamic management
In preoperatively nonsedated patients, anesthesia was induced with 3-5 μg/kg fentanyl, in 42% of the patients midazolam (0.1 mg/kg) or ketamine (1-2 mg/kg) in 18% of the patients was added if necessary followed by the application of cisatracurium (0.2-0.3 mg/kg) before endotracheal intubation. An opioid-based anesthesia (mean total dose of fentanyl equivalent 22 ± 23 μg/kg) and slightly more frequent supplemented with volatile agents in Group A (stable) patients (47% vs 25%) was used to maintain anesthesia while assuring hemodynamic stability with a target minimum mean arterial pressure of 40 mm Hg. To support hemodynamic stability, volume bolus therapy was necessary in 2/3 of all patients with no difference between the groups. However, mean transfused red blood cell volume was higher in Group B patients (6.0 ± 8.3 mL/kg (Group B) vs 2.0 ± 5.8 mL/ kg (Group A, P < .05, 95% CI 0.0-2.6). 28% (n = 52) of all patients had to be treated with vasopressors and/or positive inotropic drugs throughout the intraoperative period, including all Group B patients.
The infusion of vasoactive drugs was continued postoperatively in 44 patients. No patient required extracorporal live support.
Antiarrhythmic therapy (8% of patients) was most frequently necessary due to supraventricular tachycardia during preparation and placement of the band around the left pulmonary artery.
Adenosine and esmolol were the most commonly used drugs in this situation. More details about the different intraoperative medications are given in Table 2.

| Hemodynamic data, lactate, and cerebral nearinfrared spectroscopy
The courses of mean arterial pressure, heart rate, central venous oxygen saturation, regional cerebral saturation, and fractional cerebral tissue extraction are shown in Figure 2.

| Postoperative treatment and outcome
All patients were transferred mechanically ventilated to the PICU after the procedure. However, median postoperative ventilation time was 29 hours (IQR 38 hours), and the median postoperative length of stay in the intensive care unit was 6 days (IQR 4 days), respectively, with no difference between the groups. The overall inhospital mortality after Giessen stage I hybrid approach was 2.2% with three patients in Group A and one patient in Group B dying. Detailed information on the four deceased patients can be found in Table 3.

| D ISCUSS I ON
To the best of our knowledge, there is only little information on the anesthesiological management in patients undergoing hybrid stage I procedures despite a 30-day mortality rate that should not be underestimated. 8,9 Furthermore, no data at all exist comparing patients presenting under clinically unstable conditions, defined as inotrope therapy prior to surgery to those presenting under clinically stable conditions for the hybrid stage I approach. Our retrospective singlecenter analysis of the perioperative anesthesiological management has shown that the surgical part of GHA focusing only on bPAB is safe and effective even in patients presenting under unstable clinical conditions. An opioid-based anesthesia with low dosing of supplemented volatiles and intravenous agents was performed to ensure sufficient analgesia/anesthesia with minimal impact on hemodynamics. Over a time period of more than 16 years, there was no intraoperative death and inhospital mortality was 2.2% analyzing 185 HLHS/HLHC patients undergoing GHA. It has to be considered that the data contain a learning curve. A median heart rate of 149/min after induction of anesthesia is acceptable, but in general too high, especially for HLHS/ HLHC patients, in which the right ventricle is already pressure and volume stressed. We therefore discontinued the use of dopamine and dobutamine in 2008. In addition, we have also taken a more critical view of the indication for continuous administration of adrenaline. If one looks at the course of the mean arterial blood pressure, no clinically relevant differences are found over the time, and between the two groups. This is certainly also due to the fact that we tried to keep the MAP ≥ 40 mm Hg during anesthesia and surgery.

TA B L E 1 Patients characteristics and process times
In order to achieve this goal, 66% of the patients received at least hemoglobin between the groups, and surgical-related blood loss during bPAB could be almost excluded because not observed.
The GHA consisting of a sole performed surgical bPAB is completed by percutaneous duct stenting and atrial septostomy, in restrictive ASDs, as separate procedures. 2,3 Elective transcatheter treatment is mostly performed in extubated spontaneous breathing babies utilizing analgosedation and has been described elsewhere. 10 Considering an experienced and well-prepared team, the GHA is an easy surgical-interventional measure despite dealing with complex congenital heart defects also for newborns arriving in the hospital unstable or even in cardiogenic shock, as oftentimes observed in the past. Patients In conclusion, the described management for the stage I Giessen hybrid approach enables stable anesthesia and analgesia without compromising hemodynamics, ventilation, and perfusion even for patients with already impaired cardiac function (Group B). Rather than the preoperative status of the patient, the avoidance of extended surgery (routine placement of a reverse BT shunt, intraoperative stenting of the PDA) together with an individualized pressure and flow-oriented hemodynamic management may be accountable for the presented results.

D I SCLOS U R E S
The authors report no conflict of interest.

E TH I C S A PPROVA L
The IRB of the Justus-Liebig-University Giessen gave approval for the presented study (ref. -no. 127/15) on 13.08.2015.