An Unusual Cause of Arterial Desaturation in the Premature Infant Undergoing Patent Ductus Arteriosus Ligation

Hastings, Laura A. MD; Bushman, Gerald A. MD

Anesthesia & Analgesia:
doi: 10.1213/01.ANE.0000047088.98548.34
Author Information

Departments of Anesthesiology and Critical Care Medicine, Children’s Hospital of Los Angeles, California

October 29, 2002.

Address correspondence and reprint requests to Laura A. Hastings, MD, Departments of Anesthesiology and Critical Care Medicine, Children’s Hospital of Los Angeles, 4650 Sunset Blvd., Los Angeles, CA 90027. Address e-mail to

Article Outline

Surgical ligation of patent ductus arteriosus (PDA) in the premature neonate has become commonplace in many tertiary care neonatal intensive care units (NICU). Common co-morbidities in these patients, in addition to their high-output cardiac failure, include prematurity, respiratory distress syndrome, intraventricular hemorrhage, and renal insufficiency, which often precludes attempts at medical closure of the PDA with indomethacin. We present an unexpected and unusual cause of arterial desaturation in a premature infant undergoing thoracotomy for PDA ligation.

Back to Top | Article Outline

Case Report

A 10-day-old 560 g (24-wk gestational age) girl was scheduled for PDA ligation. Her perinatal course was complicated by intraventricular hemorrhage and hyaline membrane disease. Her PDA remained patent after an indomethacin administration. The resulting pulmonary overcirculation further compromised her existing respiratory insufficiency requiring an oscillator modality in addition to her pressure-cycled intermittent mandatory ventilation. Surgical consultation was requested, and PDA ligation was planned for the patient in the NICU.

The neonate’s baseline vital signs were blood pressure of 40/18, heart rate of 170 bpm, and pulse oximetry (Sao2) 90% on fraction of inspired oxygen 1.0 on an oscillator ventilator with four mandatory breaths/min. Physical examination, laboratory panel, and arterial blood gas measurement were unremarkable. Chest radiograph showed cardiomegaly, pulmonary overcirculation, and the appearance of hyaline membrane disease. The endotracheal tube was in good position and the lungs appropriately expanded. Monitoring included the usual noninvasive monitors and an indwelling umbilical arterial catheter. The patient was anesthetized in her warming bed in the NICU with fentanyl, ketamine, and vecuronium and turned to the right lateral decubitus position. Her vital signs, including Sao2, remained unchanged. A limited posterolateral thoracotomy was performed, and the lung was gently retracted to enable surgical dissection of the PDA. There was immediate difficulty in ventilation, as evidenced by a decrease in Sao2 to 60% and visible lack of lung movement. The endotracheal tube position was verified as the lung retractor was removed. Immediately, ventilation was restored by both mechanical and hand ventilation. Multiple attempts at different retractor placement to access the proximal descending thoracic aorta caused cessation of ventilation, despite various changes made to the ventilator settings and attempts at hand ventilation with a Jackson-Rees circuit. Removal of the lung retractors restored the ability to ventilate the patient.

At that time, surgical attention was turned to defining the anatomy of the aortic arch. With careful retraction of the left upper lobe, a right aortic arch was suspected, and a double aortic arch was identified, encircling the trachea and esophagus. The left aspect of the ring included the ductus arteriosus. Surgical examination of the aortic arch revealed further that the right posterior arch appeared 70% larger than the anterior arch and that the length of the left-sided PDA seemed to contribute to the compression of the trachea and esophagus. Because the mechanism of airway obstruction was more completely identified (vascular ring compression), the surgeons were able to make subtle changes in retractor position, which allowed adequate ventilation of the infant. After test occlusion of the PDA, the PDA and anterior left arch were clipped and divided, releasing the vascular ring. This was well tolerated by the neonate, whose blood pressure momentarily increased to 55/35 and then returned to 45/22 with a stable heart rate of 170 bpm and Sao2 of 92%. The upper lobe of the lung was then carefully reexpanded, surgical hemostasis verified, and the thoracotomy incision closed uneventfully. The patient was then returned to the supine position, appropriate ventilation and monitoring verified, and a chest radiograph obtained.

Back to Top | Article Outline


PDA ligation in the NICU has become accepted practice in many tertiary care intensive care nurseries (1). The risk of transport of the small neonate to the operating room includes hypothermia, interruption of monitoring and vascular access, accidental extubation, and difficulty with ventilation both in transport and with the anesthesia machine ventilators. Because these infants have little reserve to tolerate physiologic changes involved with the transport, it may be desirable to perform PDA ligation in the NICU with the infant in his or her warming bed. The surgical procedure itself is relatively safe. Perez et al. (2) reported 76 infants <1500 g with PDAs. The group referred for surgical management because of failure of medical management or preexisting conditions that prohibited the use of indomethacin was slightly smaller (847 g versus 997 g) and younger (gestational age 26 versus 28 weeks) compared with those treated medically. Nonetheless, there were few complications and no deaths associated with the surgery.

Anomalies of the aortic arch constitute a diverse group of conditions that may create a vascular ring around the trachea and esophagus. These anomalies are best explained by considering the normal embryologic development of the aortic arch and either persistence of various segments that should regress or regression of arch segments that should persist. In the double aortic arch, the ductus may be on the right or left, and hypoplasia or atresia of either arch may occur resulting in a ligamentum. Unlike other forms of vascular rings, persistence of the ductus arteriosus is not required to form the ring, although its size and geometry may contribute to the resulting compression (Fig. 1).

Clinically, vascular ring presentation can be variable. Whereas most are asymptomatic, those with significant tracheobronchial or esophageal obstruction present in infancy. Tracheal compression can cause stridor, recurrent respiratory infections, wheeze, or cough; esophageal compression can cause dysphagia, vomiting, or recurrent aspiration (3). These patients are frequently diagnosed by their primary care providers as having reactive airways disease, gastroesophageal reflux, or feeding intolerance. If a chest radiograph is obtained, a right aortic arch may cause the consideration of a vascular ring anomaly in the differential diagnosis. A barium swallow is the classic mode for diagnosis (4). A magnetic resonance image, computed tomography of the chest, or angiography can specifically delineate the aortic arch anatomy before surgical intervention. Often the patient has secondary tracheomalacia that may persistently cause respiratory symptoms for varying periods of time after otherwise successful surgery.

In this case, we report a premature neonate undergoing a PDA ligation where the anesthesiologist’s ability to support the patient’s ventilation requirement was dramatically affected by the unsuspected presence of a vascular ring anomaly combined with the need for modest lung retraction during thoracotomy. The recognition of the anomaly and the appropriate intraoperative decision-making to divide the ring was important in achieving a successful result from the planned PDA ligation and the avoidance of a possible additional thoracic procedure to correct persistent tracheoesophageal compression.

Back to Top | Article Outline


1. Taylor RL, Grover FL, Harman PK, et al. Operative closure of patent ductus arteriosus in premature infants in the neonatal intensive care unit. Am J Surg 1986; 152: 704–8.
2. Perez CA, Bustorff-Silva JM, Villasenor E, et al. Surgical ligation of patent ductus arteriosus in very low birth weight infants: is it safe? Am Surg 1998; 64: 1007–9.
3. Salem MR, Hall SC, Motoyama EK. Anesthesia for thoracic and cardiovascular surgery. In: Motoyama EK, Davis PJ, eds. Smith’s anesthesia for infants and children. 5th ed. St Louis, MO: CV Mosby, 1990: 508–9.
4. Backer CL, Ilbawi MN, Idriss FS, DeLeon SY. Vascular anomalies causing tracheoesophageal compression: review of experience in children. J Thorac Cardiovasc Surg 1989; 97: 725–31.
© 2003 International Anesthesia Research Society