Near-infrared spectroscopy as a predictor of clinical deterioration: a case report of two infants with duct-dependent congenital heart disease

A boy diagnosed prenatally with pulmonary atresia with an intact ventricular septum was born at our hospital at a gestational age of 37.3 weeks. After birth, the patient was admitted to the neonatal intensive care unit (NICU) with continuous positive airway pressure (CPAP) at seven cmH₂O with a maximum of 40% O₂ and infusion with prostaglandin E₁ (0.025 ug/kg/min) to maintain pulmonary circulation. Prenatal diagnosis and ductal patency were confirmed by the pediatric cardiologist. The patient appeared to be hemodynamically stable with transcutaneous oxygen saturations (SpO₂) between 70% and 90% and mean arterial blood pressures between 46 and 73 mmHg. As the patient was clinically stable, further hemodynamic parameters were not available for this case.

Four hours after birth the boy developed three apneas. The third apnea was followed by loss of cardiac output. Cardiopulmonary resuscitation was initiated immediately according to protocol. Cardiac output and heart rate, however, remained insufficient and the boy died 6 h after birth.

Autopsy confirmed the diagnosis of a pulmonary atresia with a hypoplastic right ventricle and an intact ventricular septum. Furthermore, autopsy revealed a single coronary ostium, multiple ventricular-coronary fistulas and a right ventricle dependent coronary blood flow with bridging of a branch of the right coronary artery. There were signs of recent myocardial ischemia, such as loss of cardiomyocytes. In addition, there was a left-sided open ductus arteriosus between the brachiocephalic trunk and the pulmonary artery. Based on autopsy findings, the most likely cause of death was considered to be massive myocardial ischemia which is not uncommon for this type of anatomy and physiology. Other causes such as a primary intraventricular hemorrhage or infarction cannot be ruled out completely since there was no permission for cerebral autopsy.

Baseline cerebral and renal rSO₂ were relatively low (around 40%) and decreased rapidly approximately 40 min before cardiopulmonary resuscitation was necessary (Fig. 1). This rapid decrease in cerebral and renal rSO₂was probably due to a sudden decrease in arterial oxygen saturation. All clinical parameters spontaneously recovered from this incident. Cerebral rSO₂, however, started decreasing again from approximately 40% to 15% 25 min before cardiopulmonary resuscitation was necessary, while other hemodynamic parameters did not alarm the attending staff. Within 10 min, cerebral rSO₂ had already decreased 50% from its original baseline value (40%) to approximately 20%. We cannot fully explain why cerebral rSO₂ decreased before renal rSO₂ and SpO₂.

A girl diagnosed prenatally with double outlet right ventricle, transposed great arteries, a ventricular septal defect and a hypoplastic aortic arch was born prematurely at our hospital at a gestational age of 34.9 weeks. She was admitted to the NICU with CPAP 10cmH₂O with a maximum of 40% O₂ due to respiratory distress/delayed transition. Initially, she had a wide-open ductus arteriosus. Eighteen hours after birth her clinical condition deteriorated with signs of a duct-dependent systemic circulation. Therefore, prostaglandin E₁ infusion was administered (0.025 ug/kg/min) to maintain ductal patency. During the first days after birth there was a fragile balance between pulmonary and systemic blood flow.

At the sixth day after birth, she underwent bilateral pulmonary banding with the intention of controlling an ever increasing pulmonary blood flow. Peri-operatively, hypotension was treated successfully with multiple volume expansions and noradrenalin. Postoperatively, she was on Bi-level Positive Airway Pressure (BiPAP)-assist and prior to loss of output there were no clinical signs of primary respiratory failure (e.g. inadvertent extubation). Arterial pH was 7.34, arterial saturation was 90%, pCO₂ was 4.8 kPa, hemoglobin was 8.0 mmol/l and lactate was 1.5 mmol/l. At that time, arterial blood pressure was 70/30 mmHg and she had normal diuresis. One hour later, she suddenly developed bradycardia followed by loss of cardiac output. Cardiopulmonary resuscitation was initiated immediately according to protocol. Infusion with prostaglandin E₁ was transferred from a peripheral venous site to a central line and was increased in dosage. An echocardiogram suggested an open ductus arteriosus, revealed a poor contractility and an inadequate cardiac output. After 40 min of cardiopulmonary resuscitation, the sternum was re-opened. The heart, however, was almost non-contractile and the girl died 7 days after birth.

Autopsy confirmed the diagnosis and concluded that the lumen of the ductus arteriosus was small with a diameter of less than three millimeter. Whether this was too small to maintain systemic circulation and whether prostaglandin E₁ infusion was inadequate at the time of collapse, remains unknown. Furthermore, autopsy revealed an immature lung parenchyma, atelectasis of part of the basal lobes and small inflammatory areas in both lungs. These pulmonary findings might also have contributed to the cardiopulmonary collapse.

We observed an increase in cerebral rSO₂ and a decrease in renal rSO₂ 1 h before loss of output (Fig. 2). Renal rSO₂ returned to baseline, while cerebral rSO₂ remained approximately 10% above baseline. This incident might have been caused by a closing ductus arteriosus due to insufficient Prostaglandin E₁ administration or an insufficient response to adequate Prostaglandin E₁ administration. Especially the rapidly decreasing renal rSO₂ was indicative of a closing ductus arteriosus. Due to the coarctation of the aorta, renal blood flow was dependent on an open ductus arteriosus. The increase in cerebral rSO₂ was less easy to explain, but could be caused by an increase in cerebral blood flow. Due to the closing ductus arteriosus, the descending aorta might have become narrower, leading to an increased arterial pressure above the coarctation and subsequently increased cerebral blood flow. Thirty minutes before cardiopulmonary resuscitation, cerebral rSO₂ started decreasing from 55% to approximately 20% and renal rSO₂ decreased from 40% to 15%. Renal rSO₂ decreased more rapidly compared with cerebral rSO₂ and approximately 15 min before loss of output, renal rSO₂ was less than 50% from its original baseline value. The more rapid decrease of renal rSO₂ in comparison with cerebral rSO₂ also supports the theory of insufficient systemic circulation due to a closing ductus arteriosus. Apart from a slowly increasing heart rate from 160 bpm to 180 bpm, hemodynamic parameters such as SpO₂ and arterial blood pressure remained stable until several minutes before cardiopulmonary resuscitation was necessary (Fig. 2).