Use of Extracorporeal Membrane Oxygenation and Cerebral Oximetry in a Stage 1 Norwood Repair for Hypoplastic Left Heart Syndrome

Abstract

The role of extracorporeal membrane oxygenation (ECMO) and its indications in stage I Norwood palliation are controversial. The decision to initiate ECMO and its timing remains difficult with no definitive cut-off points or evidence-based guidelines. It varies on a case-to-case basis. We report a case where the use of ECMO was beneficial after stage I Norwood palliation with severe ventricular dysfunction. The systemic-to-pulmonary artery shunt was kept open to balance the systemic and pulmonary circulations. Cerebral oximetry can be useful as an additional monitoring modality to guide management, monitor cerebral perfusion, and help detect cerebral steal.

INTRODUCTION

Hypoplastic left heart syndrome (HLHS) is a complex congenital malformation; features of which are characterized by underdevelopment of the left-sided structures of the heart.[¹] Morphological features of HLHS include a combination of wide variations such as hypoplasia of the left ventricle, mitral valve and aortic valve stenosis or valvular atresia, and hypoplasia of the ascending aorta.[²] Early diagnosis in the neonatal period using echocardiography with the use of prostaglandins for ductal patency led to the initiation of palliative—staged reconstructive procedures for HLHS described by Norwood et al.[³] The Norwood procedure is a stage 1 palliative reconstructive surgery which is the construction of an unobstructed systemic blood flow from the right ventricle to the aorta, and the establishment of optimal pulmonary blood flow through a systemic-to-pulmonary artery shunt.[^1,3,4] Extracorporeal membrane oxygenation (ECMO) is an important circulatory assist for refractory cardiopulmonary dysfunction, but its role and indications after stage 1 Norwood palliation are debatable.[⁵] Cerebral oximetry monitoring with near-infrared spectroscopy (NIRS) allows continuous, quantitative, and non-invasive assessment of regional cerebral oxygen saturation to predict neurological injury or hypoxic risk and is widely being used perioperatively in neonatal heart surgery.[⁶]

CASE HISTORY

A 12-day-old neonate weighing 2.9 kg was diagnosed with congenital heart disease on day 2 of life; 2D echocardiography was suggestive of HLHS, severe mitral stenosis, mitral valve annulus (MVA) 0.32 cm, small aortic valve (0.55 cm) with sub-valvar narrowing, hypoplastic transverse aortic arch, large patent ductus arteriosus, normal tricuspid valve with mild regurgitation, large subaortic ventricular septal defect, malaligned interatrial septum, non-restrictive pulmonary venous flow to the left atrium and small atrial septal defect with left-to-right shunt. The child was stabilized preoperatively in the intensive care unit (ICU), electively posted for stage 1 Norwood palliation on ICU day 2. Anesthesia induction was uneventful, opioid-based, and maintained with inhalational agents and muscle relaxant. Standard monitoring protocol with electrocardiography, pulse oximetry, capnometry and NIRS, two invasive arterial lines (one in the right upper limb and the other in the lower limb), and central venous line (femoral vein) were put.

Median sternotomy was performed and aorta-right atrium cardiopulmonary bypass (CPB) was instituted. Stage 1 Norwood procedure was done; right modified Blalock Taussig (RMBT) shunt of 3.5 mm was put. The patient had two pump runs in view of desaturation and ST-T changes while weaning off CPB- cardiopulmonary bypass lie of the shunt (meaning direction of placement of the shunt). Total CPB duration was 188 + 15 minutes, with aortic cross clamp of 69 minutes and selective innominate perfusion for 69 minutes. Last intraoperative arterial blood gas showed some respiratory acidosis with pH of 7.24, partial pressure of CO_2: 68.5 mmHg, partial pressure of Oxygen: 43.2 mmHg, saturating: 71% with lactates of 7.52. NIRS baseline value (before induction) was 52 and 56 before shifting to ICU [Figure 1].

Figure 1: Patient on ECMO support with invasive and non-invasive monitoring, cerebral oximetry with near infrared spectroscopy

Patient was shifted to ICU with open stented chest on high inotropic support (infusions of adrenaline 0.1 mcg/kg/min, noradrenaline 0.05 mcg/kg/min, milrinone 0.5 mcg/kg/min, calcium 100 mg/h) and ventilatory settings; FiO2: 60%, rate: 60/min. Immediate ICU course was uneventful; hemodynamic parameters maintained; acceptable blood gases with pH of 7.35, partial pressure of CO_2: 45 mmHg, partial pressure of O_2: 43.5 mmHg, lactates of 4.56 showing decreasing lactate trend. NIRS monitoring was continued in the immediate post-operative period. NIRS baseline value (on ICU arrival) was 54.

Six hours into the post-operative course, the patient had sudden hypotension with bradycardia. Direct cardiac massage with bolus doses of adrenaline was given; return of spontaneous circulation was achieved within a minute. Due to low saturation and persistent hypotension despite high inotropes, emergency CPB was initiated, RMBT shunt revision was done (from 3.5 to 4 mm), but unable to wean off CPB due to severe biventricular dysfunction, the patient was put on support bypass which was transitioned to veno-arterial (VA) ECMO. VA ECMO was initiated at a flow rate of 0.45 L, 2,300 rpm, sweep gas flow rate of 0.3 L/min, FiO2 30%, and perfusion pressure of 40-48 mmHg was maintained. The RMBT shunt was kept open and the heart was kept pulsatile with ejections. Inotropic support was optimized to the minimum and near-full ventilation FiO2 50, rate: 30/min, 85-90% saturation, target PO2 of 60-70, serial arterial blood gases, and electrolytes were measured [Figures 2 and 3].

Figure 2: Chest X-ray view showing ECMO venous and arterial cannula, endotracheal tube with bilateral drains in situ

Figure 3: Chart A showing values of partial pressure of carbon dioxide (PaCO2), partial pressure of oxygen (PaO2), mean arterial pressure (MAP), Heart rate (HR), near-infrared spectroscopy (NIRS), pulse oximetry (SPO2). Chart B shows values of ECMO flow rates

Cerebral oximetry was used as an additional neuromonitoring tool. NIRS values were monitored throughout ECMO ranging between 48 and 58, the average value of 54. Changes in NIRS values from baseline were managed with optimizing ECMO settings: flow rate, ventilatory settings, and preload/afterload mismatch. There were three instances of a drop in the NIRS values (more than 20% from baseline), twice the ECMO flows were increased (26 and 36 h postoperatively), and in one instance, we modified the preload and ventilatory settings (48 h postoperatively) [Figure 3]. The patient was successfully decannulated after VA ECMO of 50 hours 40 minutes. The chest was closed on post-operative day (POD) 6, extubated on POD 10. Post-operative course was complicated with infection for which sensitive antibiotics were used. The patient got discharged on POD 56.

DISCUSSION

The role of ECMO and its indications in stage I Norwood are controversial and have been reported less from developing countries like ours. We only have a few reports of the Norwood repair for HLHS with the use of ECMO postoperatively. Indications to put on ECMO are not clearly defined ranging from inability to separate from CPB requiring support for prolonged duration, hypoxia, hypotension, low cardiac output state and cardiac arrest.[⁵] High-risk factors requiring ECMO support derived from previous studies attribute to low birth weight, neonatal age group, complex cardiac anatomy, increased CPB duration and increased need for inotropes, turbulent intraoperative and immediate post-operative course.[⁷] Previous reports.[^5–9-] have studied the use of ECMO post-Norwood palliation electively and non-electively with varying results affecting morbidity and mortality. Variability in survival post-ECMO also depends on the differences in patient selection, the timing of ECMO deployment and overall patient management and expertise. While there is no common consensus regarding shunt management during ECMO, J Jaggers et al.[¹⁰] have published a study regarding shunt management post-cardiotomy support. Even though current practice suggests to occlude the shunt completely or partially to limit pulmonary blood flow, taking some insights from the said study,[¹⁰] and as a team, our decision was to keep the shunt open to perfuse the lungs and maintain high ECMO flow rates with optimal ventilation to support both pulmonary and systemic circulation. We did not experience any problems related to pulmonary overcirculation. The high ECMO flow rates were also able to avoid any phenomenon of cerebral and coronary steal, which was aided by NIRS monitoring in the ECMO period. As shown in Figure 3, during the period of suboptimal NIRS readings, the ECMO flow rates were adjusted accordingly to balance the circulation. Initiation of ECMO and its timing in this subset remains a difficult subject of debate with no definitive cut-off points or evidence-based guidelines; its practice varies on a case-to-case basis. Hence, we conclude that VA ECMO can be beneficial in cases of transient but severe ventricular dysfunction and sudden circulatory collapse after stage I Norwood palliation. The use of cerebral oximetry guided by NIRS will help in the overall management and provide an additional tool to monitor cerebral perfusion during the period of ECMO, especially when the systemic-to-pulmonary shunt is kept open.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form the patient (s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.