Abstract
The use of prostaglandin infusion to maintain patency of the ductus arteriosus in patients with critical coarctation of the aorta (CoA) to support systemic circulation is the standard of care. However, pulmonary overcirculation resulting from a patent ductus arteriosus in patients with critical CoA is not well described in the literature. We report two cases of critical CoA that required invasive measures to control pulmonary blood flow before surgical repair of the CoA. Both patients had signs of decreased oxygen delivery, hyperlactatemia, and systemic to pulmonary flow via the ductus arteriosus. One patient required surgical pulmonary artery banding and the second patient underwent pulmonary flow restrictor device placement for the control of pulmonary blood flow. A rapid improvement in oxygen delivery and normalization of lactate levels were observed after control of pulmonary overcirculation. Both patients underwent successful surgical repair of the coarctation A and were discharged home.
Keywords
Introduction
Coarctation of the aorta (CoA) is the narrowing of the proximal thoracic aorta and is seen in 2.5 to 4 per 10 000 live births, 1 comprising about 2.6% of all congenital heart disease (CHD). 2 Critical coarctation of the aorta is defined as a CoArequiring ductal patency for adequate systemic perfusion. This can be a neonatal emergency characterized by obstructive cardiogenic shock with increased left ventricular afterload. The use of prostaglandin to maintain patency of the ductus arteriosus to support the systemic circulation before surgical repair is well-documented and is the standard of care. 3 However, pulmonary overcirculation resulting from the ductus arteriosus in patients with critical CoA is uncommon and is not well described. We report two cases of critical CoA that required invasive measures to control pulmonary blood flow before surgical repair of the CoA. The requirement of parental consent to publish this case report was waived as per our institutional review board in accordance with its policy.
Case 1
A 3.31 kg full-term neonate with a prenatal diagnosis of hypoplastic aortic arch was born via induced vaginal delivery to a 41-year-old primigravida mother. Prostaglandin was initiated in the delivery room given suspected ductal dependent systemic circulation. Postnatal ultrasound at 13 h of life showed a critical aortic coarctation with a discrete narrowing at the isthmus, hypoplastic transverse aortic arch (2.8 mm, z score −3.1), and a large patent ductus arteriosus with bidirectional flow. On hospital day (HD) #3, the patient developed tachycardia and lactic acidosis despite a patent ductus arteriosus (PDA) on echocardiogram. A sepsis work-up returned positive for Escherichia coli meningitis for which antibiotics were initiated. On HD #9 the patient had acute worsening of perfusion, tense abdominal distension, and increasing lactate levels with a peak lactate level of 11.4 in the presence of normal biventricular function and a PDA on echocardiogram (Figure 1A and B and Figure 2). An emergent laparotomy revealed a well-perfused small and large bowel without evidence of any intestinal necrosis or perforation. Given that the patient had persistent lactic acidosis despite medical management and fluid resuscitation, a provisional diagnosis of pulmonary overcirculation with systemic circulation compromise was made. The patient underwent bilateral pulmonary artery (PA) banding to control pulmonary blood flow and improve systemic perfusion. The postoperative course was complicated by supraventricular tachycardia, which resolved with medical management. Postoperative echocardiogram revealed well-positioned PA bands with flow accelerations across both branch pulmonary arteries and a new echogenic mass on the mitral valve, concerning for endocarditis. The patient continued on broad-spectrum antibiotics and improved with normalized lactate levels by postoperative day 2. On HD #17, the patient underwent bilateral branch PA band removal, ductal ligation, and aortic arch repair. Postoperative echocardiogram revealed a patent arch with normal aortic doppler pattern and resolution of the previous mitral valve vegetation. Recovery was uncomplicated, and the patient remained hospitalized for completion of a six-week course of antibiotics for presumed endocarditis and E coli meningitis after which he was discharged home.

(A) 2D and color echo image showing discrete coarctation of the aorta. (B) Abdominal aortic Doppler showing reversal of flow in the descending aorta suggestive of flow through the ductus arteriosus from the thoracic aorta to pulmonary artery.

Case 1. Serum lactate trend.
Case 2
A 2.33-kg male was born at 37 weeks gestation with intrauterine growth restriction. He was born at an outside hospital via a Cesarean section for nonreassuring fetal heart tracing. The baby was initially admitted to the neonatal intensive care unitfor persistent hypoglycemia but remained stable on room air. A postnatal echocardiogram was done due to a lack of optimal visualization of the heart during the prenatal course that demonstrated a hypoplastic distal transverse arch (3.5 mm, z score −2.95), severe discrete CoA (2.2 mm, z score −2.52), a large PDA with bidirectional shunting, a small anterior muscular ventricular septal defect (VSD), and normal biventricular systolic function. Prostaglandin infusion was initiated, and the patient was transferred to our hospital. Soon after arrival, the patient developed lactic acidosis. An echocardiogram showed normal biventricular function and a PDA.He initially responded to fluid resuscitation and milrinone but showed decompensationagain on HD #2 with worsening respiratory and metabolic acidosis. A chest x-ray was concerning for pulmonary congestion, and an echocardiogram showed a large ductus arteriosus with bidirectional flow and normal ventricular systolic function. The patient was intubated, diuresis was initiated, and empiric broad-spectrum antibiotics for presumed septic shock were initiated. On HD #4, the baby developed worsening lactic acidosis with hypotension. A sepsis evaluation returned negative. A head ultrasound showed bilateral grade 2 intraventricular hemorrhages. Despite inotropic support, fluid resuscitation, and antibiotics, the patient continued to have uptrending lactates with a peak lactate level of 14.4 (Figure 3A and B and Figure 4). A diagnosis of pulmonary overcirculation was considered, and the patient was taken to the cardiac catheterization laboratory which showed evidence of excessive pulmonary blood flow by angiography and elevated right ventricular end diastolic pressure of 12 mm Hg. A pulmonary flow restrictor (PFR) device (Micro Vascular Plug MVP-7Q®, Medtronic) was placed in each PA. Post-procedural echocardiogram showed mostly antegrade flow across the aortic arch. The patient demonstrated gradual clinical improvement with resolution of lactic and metabolic acidosis. The post-procedural course was complicated by a grade 3 germinal matrix hemorrhage that was managed conservatively and planned anticoagulation for the PFR devices could not be commenced. Four weeks after PFR placement, the patient underwent an uncomplicated aortic arch repair with pulmonary homograft, removal of PFRs, and PDA ligation and division. Intraoperatively, a significant degree of neointimal response was seen in both PAs due to the PFR devices, but the devices were successfully extracted in toto. Postoperative echocardiogram showed normal flow in the aortic arch, a widely patent right pulmonary artery, mild flow acceleration in the midportion of the left pulmonary artery, and normal biventricular function. The remainder of the postoperative course and recovery were uneventful, and the patient was discharged home on postoperative day 10.

(A) 2D and color echo image showing discrete coarctation of the aorta. (B) Abdominal aortic Doppler showing reversal of flow in the descending aorta suggestive of flow through the ductus arteriosus from the thoracic aorta to pulmonary artery.

Case 2. Serum lactate trend. Abbreviation: PA, pulmonary artery
Discussion
We report two neonates with critical CoA who had evidence of pulmonary overcirculation requiring interventions to manage pulmonary overcirculation prior to surgical repair of CoA. To our knowledge, this phenomenon has not been described well and we could not find any similar reports of its description. Although PA banding in patients with CoA has been reported in patients with hemodynamically significant VSDs to control excessive pulmonary blood flow, to our knowledge it has not been reported in patients with isolated CoA. 4
Hyperlactatemia, or elevated lactate levels, is described as type A where tissue hypoxia is present and type B in the absence of tissue hypoxia. 5 Hyperlactatemia in patients with critical CoA is not uncommon and is secondary to decreased systemic perfusion (Type A) leading to poor oxygen delivery, intestinal ischemia and/or necrotizing enterocolitis, or sepsis. It is important to consider common causes of lactate elevation in a patient with critical CoA, such as ductal closure or constriction, ventricular dysfunction, sepsis, and necrotizing enterocolitis. However, one must also consider systemic steal resulting from left to right flow via the PDA resulting in pulmonary overcirculation at the expense of systemic circulation. Although a PDA is critical in maintaining systemic circulation in patients with critical CoA, the patent ductus can also be responsible for systemic steal in the setting of decreasing pulmonary vascular resistance in the first few days after birth as evident in our cases. We agree that the severe degree of derangement from only pulmonary overcirculation is highly unusual. Most patients with CoA and PDA would not have clinical derangement similar to what we have described. However, these two cases seek to highlight the potential for this phenomenon in some patients. While it is not entirely clear why some patients quickly slip into pulmonary overcirculation and shock, this may be related to critical interplay between systemic and pulmonary vascular resistance. However, we did undertake an extensive work-up to rule out secondary etiologies such as intra-abdominal pathology, sepsis, and echocardiogram to confirm ductal patency and evaluate ventricular function which were all reassuring. The hyperlactatemia and hemodynamic derangement resulted from systemic steal and ischemia resulting from pulmonary overcirculation and not from obstruction to systemic perfusion which is often the case in patients with critical CoA. If medical measures to control pulmonary blood flow and to maintain systemic oxygen delivery are insufficient, as in our cases, surgical or nonsurgical intervention to limit pulmonary blood flow may be required to stabilize the patient prior to definitive surgical repair.
The dilemma of surgical intervention on critically ill infants with CHD is challenging. Although the patient is critically ill secondary to underlying CHD, surgical repair in the phase of acute hemodynamic instability may have undesirable consequences. The presence of preoperative hemodynamic instability with shock, acidosis, need for mechanical circulatory, and respiratory support are known risk factors associated with increased morbidity and mortality in children undergoing surgical repair of CHD. 6 In addition, preoperative hyperlactatemia in infants undergoing surgery for congenital cardiac disease is not only an early marker of morbidity and mortality but also associated with neurodevelopmental impairment in later life.7,8 Hence, hemodynamic optimization prior to surgical repair is imperative in critically ill infants with congenital heart disease. Our institutional policy for neonates with critical arch obstruction is to perform the surgical repair after the patient's physiology has stabilized with restoration of ductal patency. In patients with prenatal diagnosis and postnatal confirmation of critical CoA or hypoplastic arch, prostaglandin infusion is initiated immediately after birth. In patients with a new postnatal diagnosis or late diagnosis, prostaglandin infusion is initiated to restore ductal patency and establish systemic perfusion to allow recovery from organ dysfunction if any. Hence, most patients with critical CoA undergo surgical repair in the first week of life. We acknowledge that the delay in surgical repair of these two patients is outside of our normal management protocol and not the standard of care at our institution. Both of these cases had active medical diagnoses such as E coli meningitis in the first case and intraventricular hemorrhage in the second case. Hence, the definitive surgical repair in these patients was delayed to allow for medical recovery and hemodynamic stabilization.
The use of PFRs in controlling pulmonary blood flow is novel and is being increasingly reported.9–11 Schaffner et al 9 used it in a patient with truncus arteriosus with hypoplastic right ventricle and severe truncal valve stenosis while waiting for a heart transplant and described its use for 170 days with uneventful removal at the time of cardiac transplant. Similarly, two short case series involving six patients 10 and 17 patients 11 with hypoplastic left heart syndrome and its variants have described the use of PFRs as part of a hybrid stage 1 procedure documenting their safety and efficacy. However, to our knowledge, the use of PFRs in patients with critical CoA for temporary control of pulmonary blood flow has not been described before. Cardiac surgery in critically ill patients with single/biventricular physiology may be temporarily deferred if they are too young, being treated for an acute medical condition, or unstable for major cardiac surgery. As in our case, the use of PFRs could be considered in such patients to control the pulmonary blood flow when pulmonary overcirculation is of concern. It is not our institutional policy to perform balloon angioplasty of native neonatal CoA as it is associated with a high recurrence rate requiring reintervention, higher incidence of aortic aneurysm, and high incidence of procedural morbidities. 12 We, however, consider balloon angioplasty for patients with recoarctation after surgical repair and CoA diagnosed in older children and adolescents.
In conclusion, patients with ductal dependent systemic circulation should be monitored closely for adequacy of systemic circulation as compromised systemic cardiac output can be seen even with ductal patency. Additional measures of controlling pulmonary blood flow may be needed in these patients if they are unresponsive to medical measures for the control of excessive pulmonary blood flow.
Footnotes
Abbreviations
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
