Key words: Airway: obstruction. Anesthesia: pediatric. Complications: respiratory. Lung: pulmonary artery compression. Mediastinal mass. Trachea: compression.

THE patient with an anterior mediastinal mass presents the anesthesiologist with a formidable challenge. Severe respiratory and cardiovascular complications, including the inability to resuscitate, can ensue on induction of general anesthesia. We describe the anesthetic management of a child with a large dynamic vascular malformation located in the neck and mediastinum who underwent percutaneous sclerotherapy.

A 9-yr-old girl was admitted to our institution for xsclerotherapy of a large mediastinal vascular malformation. At age 2 yr, the patient was noted to have a small mass of the left neck, initially believed to be a hemangioma, and was treated with steroids with no effect. She remained asymptomatic until age 7 yr, when she was evaluated for dyspnea. Workup at that time suggested the mass was an arteriovenous malformation that was intermittently compromising respiration. Magnetic resonance imaging and magnetic resonance angiography revealed an extensive, lobulated mass in the left neck and upper chest, extending from the angle of the mandible through the neck and thoracic inlet into the mediastinum below the aortic arch (Figure 1and Figure 2). The trachea was mildly compressed and displaced anteriorly and to the right. The left carotid and left vertebral arteries appeared encased by the mass. Extension of the vascular malformation into the retropharyngeal region of the hypopharynx was noted. Initial angiography showed the predominant blood supply to the malformation to be intimately associated with a feeding vessel to the anterior spinal artery, with venous drainage to the paraspinal veins and azygous system. Because of the risks of intervention, the patient was treated conservatively for 2 yr. During the 6 months before admission, however, she began to experience increasingly frequent episodes of dyspnea and stridor associated with neck swelling. There was progressive orthopnea, and she required three pillows for sleep. Pulmonary function tests demonstrated a mild obstructive impairment (Table 1). Attempted sclerosant injection at another hospital was aborted after the lungs became difficult to ventilate after intravenous induction of anesthesia with thiopental and succinylcholine. It was noted at that time that the neck swelled with positive pressure ventilation, and only the epiglottis could be visualized on direct laryngoscopy. With return of spontaneous breathing, the neck swelling improved, and the patient recovered uneventfully. She was referred to our institution.

Figure 1. A photograph of the child was taken at 7 yr of age, before treatment, demonstrating the anterior neck and supraclavicular soft-tissue mass.

Figure 1. A photograph of the child was taken at 7 yr of age, before treatment, demonstrating the anterior neck and supraclavicular soft-tissue mass.

Close modal

Figure 2. Magnetic resonance imaging of the neck and mediastinum before treatment. (A) Coronal T1-weighted, gadolinium-enhanced image through the level of the trachea demonstrates the extensive enhancing lesion surrounding the left carotid artery (LCA) and displacing the trachea (T) to the right. (B) Axial T2-weighted image demonstrating the anteroposterior extent of the malformation at the level of the lower neck, again showing the mass to surround the LCA and left vertebral artery (LVA) and displace the trachea to the right.

Figure 2. Magnetic resonance imaging of the neck and mediastinum before treatment. (A) Coronal T1-weighted, gadolinium-enhanced image through the level of the trachea demonstrates the extensive enhancing lesion surrounding the left carotid artery (LCA) and displacing the trachea (T) to the right. (B) Axial T2-weighted image demonstrating the anteroposterior extent of the malformation at the level of the lower neck, again showing the mass to surround the LCA and left vertebral artery (LVA) and displace the trachea to the right.

Close modal

Table 1. Improvement in Pulmonary Function Tests after Sclerotherapy

Table 1. Improvement in Pulmonary Function Tests after Sclerotherapy
Table 1. Improvement in Pulmonary Function Tests after Sclerotherapy

Examination revealed a well developed and well nourished girl weighing 29.2 kg and 135 cm in height, with a palpable nonpulsatile soft-tissue mass of the left neck and supraclavicular area. The mass distended with the Valsalva maneuver and was easily compressible. Audible stridor occurred on lying flat, and the patient complained of dyspnea when supine. Facial edema was not present. Supine and upright SpO2was 98-100% on room air.

The patient was brought to the interventional radiology suite and noninvasive monitors were placed. Peripheral intravenous access was established, and anesthesia was induced with halothane, nitrous oxide, and oxygen with the head elevated 30 degrees. Spontaneous respiration was maintained with 5 cmH2O positive end-expiratory pressure, controlled with the adjustable pressure-limiting valve, and the trachea was intubated under deep inhalational anesthesia without the use of muscle relaxants. Direct laryngoscopy with a Macintosh 2 blade revealed displacement of the supraglottic larynx and the laryngeal inlet to the right. Following the translaryngeal passage, the tracheal tube was advanced to within 1 cm of the carina. Attempts at controlled respiration resulted in engorgement of the neck mass associated with a decrease in oxygen saturation from 100% to 97%. Spontaneous respiration was therefore resumed and maintained throughout the procedure. Additional vascular access was established, and a radial arterial catheter was inserted. Percutaneous embolization of the neck and supraclavicular mass was performed by injection of 30 ml of 100% ethanol, 4 ml of 3% sodium tetradecyl sulfate (Sotrecol) and placement of 11 platinum coils. One hundred forty milliliters of blood was withdrawn from the lesion. Anesthesia was maintained with isoflurane, nitrous oxide, and titrated doses of fentanyl. Two hours into the procedure, the patient's SpO2began to fall; an arterial blood gas on 100% oxygen revealed a pH of 7.19, pCO2of 52 mmHg, and PO2of 115 mmHg, with a base excess of -9. The procedure was quickly concluded and the patient transported to the intensive care unit with the tracheal tube in place. Sedation was maintained overnight with isoflurane, ketamine, and droperidol, with continued spontaneous breathing. The following morning, the neck mass was noted to be swollen but firm. A trial of positive pressure ventilation was well tolerated and did not result in distention of the neck mass or arterial desaturation. The patient was returned to the interventional radiology suite that day for a second course of sclerotherapy, which proceeded uneventfully.

The patient's trachea remained intubated in the intensive care unit after her sclerotherapy to allow the edema of her neck to resolve. Her trachea was extubated uneventfully on the 4th postoperative day. Two months later, she was readmitted for a further course of sclerotherapy to treat the deeper mediastinal portion of the venous malformation. Anesthetic management during the 4.5-h procedure was uneventful, and she was able to tolerate positive pressure ventilation without difficulty. The patient returned for a final sclerotherapy session 7 months after the first one, with significant interval improvement: She was able to sleep normally with one pillow, played basketball with her friends, and had a corresponding improvement in her pulmonary function tests (Table 1). Followup magnetic resonance imaging revealed thrombosis and a decrease in size of the lesion, and physical examination showed a marked reduction in the external dimensions of the mass.

Our patient's symptoms of stridor and orthopnea were suggestive of significant airway narrowing. Many patients with an anterior mediastinal mass will present with cough, dyspnea, stridor, or orthopnea. Absence of symptoms, however, does not assure a trouble-free anesthetic. [1-3]Several authors have suggested that preoperative pulmonary function tests performed in the supine position may be helpful; flow-volume curves showing limitation of expiratory flow may presage airway collapse with general anesthesia. [4,5]Measurement of tracheal cross-sectional area by computed tomography may be helpful: Patients with tracheal narrowing greater than 50% appear to be at significantly greater risk for airway obstruction. The combination of pulmonary function test abnormalities (increased residual volume and decreased peak expiratory flow ratios in supine compared to sitting position) and computed tomography may serve to enhance awareness of the patient with a mediastinal mass and significant airway compromise for whom general anesthesia is a serious risk.

We chose to induce anesthesia with inhalational agents and continue spontaneous ventilation, as recommended by Ferrari et al. [6]In addition, we maintained 5 cmH2O positive end-expiratory pressure to prevent pharyngeal collapse and provide a modest defense against the loss of functional residual capacity that accompanies anesthetic induction. [7,8]General anesthesia in the supine patient can exacerbate extrinsic airway compression by reducing chest wall diameter and promoting cephalad displacement of the diaphragm, thus reducing overall thoracic volume and decreasing the space for the mass. [3]By increasing intrathoracic pressure, positive pressure ventilation may have resulted in a blood volume shift from the intrathoracic to the extrathoracic portion of the vascular malformation, creating a compressive effect on extrathoracic airway structures. This hypothesis is similar to the intrathoracic fluid shifting found by Warner et al. in their evaluation of chest wall motion [9]and is supported by lack of recurrence in this patient after the initial embolization of the neck portion of the mass. It is not uncommon to have to place patients with a mediastinal tumor in semi-Fowler's or sitting position to decrease their orthopnea and maintain airway patency; often these patients choose such positions for comfort, as did ours. By performing tracheal intubation under deep inhalational anesthesia without the use of muscle relaxants, positive pressure ventilation was avoided, and a more normal transpulmonary pressure gradient was maintained, which may have decreased the shifting of blood to the extrathoracic portion of the malformation. In addition, airflow through conducting airways may be better maintained with spontaneous ventilation in such patients. [10]Although not an issue in this case, respiratory problems can occur on emergence from anesthesia as a result of edema or surgical manipulation.

Hypoxemia at the end of the procedure (SpO2of 84% with an inspired oxygen concentration of 0.4) may have been related to progressive atelectasis with shunting, and acutely exacerbated by movement of fresh thrombi from the malformation into the central circulation. This was accompanied by mild hypotension, although the end-tidal carbon dioxide did not change. Treatment with 100% Oxygen2and lightening of the anesthetic promoted resolution within several minutes. When administered in the venous circulation, the vasospasm effects of sclerotherapy are immediate, [11]locally restricted, [12,13]and involve the vascular endothelium and platelets in the area of injection, producing a locally hypercoagulable state. [14]Although pulmonary vascular effects generally are believed to be benign after sclerotherapy with absolute alcohol, [15]cardiovascular collapse, preceded by hypoxemia and hypercarbia, has been reported in four patients undergoing ethanol embolization or sclerotherapy. In these cases, the mechanism was believed to involve pulmonary vasoconstriction after sclerosant injection. [16].

Decreased cardiac output and/or blood pressure, even when ventilation is apparently adequate, is also a risk in patients with a mediastinal mass. [16-18]The etiology of these hemodynamic changes is considered to be compression of the great vessels and right atrium. Studies in animals have confirmed that cardiac output decreases in the presence of an anterior mediastinal mass. This appears to be the result of an increase in right ventricular afterload with corresponding right ventricular dysfunction and appears independent of muscle paralysis or spontaneous versus mechanical ventilation. [19]Others have suggested that, with positive pressure ventilation, pulmonary arterial and venous flow might decrease, so that blood volume in a mass with intrathoracic and extrathoracic components might be expected to shift from the intrathoracic to the extrathoracic space. [20]With our patient breathing spontaneously for the entire anesthetic, perhaps the preservation of normal vascular transit time was the most important issue in minimizing the expansion of the extrathoracic portion of the malformation.

This is the first reported case to our knowledge of a dynamic anterior mediastinal mass causing significant adverse cardiorespiratory changes with induction of anesthesia. Although vascular malformations of the mediastinum are relatively uncommon, comprising only 5% of pediatric mediastinal tumors in one large series, [21]their dynamic nature makes them particularly hazardous for general anesthesia. Advance planning should include consideration for special aspects of anesthetic management of patients with vascular mediastinal masses undergoing sclerotherapy, such as maintenance of spontaneous ventilation, anticipation of the vascular and histologic consequences of sclerosing agents used in the central circulation, and close cooperation with other members of the operating team.

Azizkhan RG, Dudgeon DL, Buck JR, Colombani PM, Yaster M, Nichols D, Civin C, Kramer SS, Haller JA: Life-threatening airway obstruction as a complication to the management of mediastinal masses in children. J Pediatr Surg 1985; 20:816-22.
Bray RJ, Fernandes FJ: Mediastinal tumor causing airway obstruction in anaesthetised children. Anaesthesia 1982; 37:571-5.
Pullerits J, Holzman RS: Anaesthesia for patients with mediastinal masses. Can J Anaesth 1989; 36:681-8.
Neuman GG, Weingarten AE, Abramowitz RM, Kushins LG, Abramson AL, Ladner W: The anesthetic management of the patient with an anterior mediastinal mass. ANESTHESIOLOGY 1984; 60:144-7.
Prakash UBS, Abel MD, Hubmayr RD: Mediastinal mass and tracheal obstruction during general anesthesia. Mayo Clin Proc 1988; 63:1004-11.
Ferrari LR, Bedford RF: Anterior mediastinal mass in a pregnant patient: Anesthetic management and considerations. J Clin Anesth 1989; 1:460-3.
Rehder K: Anesthesia and the mechanics of respiration, Effects of Anesthesia. Edited by Covino BG, Fozzard HA, Rehder K, Strichartz G. Bethesda, American Physiological Society, 1985, pp 91-106.
Marsh HM, Southorn PA, Rehder K: Anesthesia, sedation, and the chest wall (review). Int Anesthesiol Clin 1984; 22:1-12.
Warner DO, Krayer S, Rehder K, Ritman EL: Chest wall motion during spontaneous breathing and mechanical ventilation in dogs. J Appl Physiol 1989; 66:1179-89.
Sibert KS, Biondi JW, Hirsch NP: Spontaneous respiration during thoracotomy in a patient with a mediastinal mass. Anesth Analg 1987; 66:904-7.
Perchuk E, Tafreshi M, Radow A, Smulewicz JJ: Sclerotherapy of varicose veins: Phlebographic demonstration of immediate venous spasm. NY State J Med 1977; 77:204-8.
Williams RA, Wilson SE: Sclerosant treatment of varicose veins and deep vein thrombosis. Arch Surg 1984; 119:1283-5.
Cho KJ, Williams DM, Brady TM, Weiss CA, Forrest ME, Oke EJ, Griffin DJ: Transcatheter embolization with sodium tetradecyl sulfate: Experimental and clinical results. Radiology 1984; 153:95-9.
Jacobson BF, Franz RC, Hurly EM, Norman GL, Becker P, Myburgh JA, Mendelow BV: Mechanism of thrombosis caused by sclerotherapy of esophageal varices using sodium tetradecyl sulphate. Surg Endosc 1992; 6:4-9.
Sarin SK, Sethi KK, Nanda R: Pulmonary hemodynamic changes after intravariceal sclerotherapy with absolute alcohol. Gastrointest Endosc 1988; 34:403-6.
Keon TP: Death on induction of anesthesia for cervical node biopsy. ANESTHESIOLOGY 1981; 55:471-2.
Halpern S, Chatten J, Meadows AT, Byrd R, Lange B: Anterior mediastinal masses: Anesthesia hazards and other problems. J Pediatr 1983; 102:407-10.
Levin H, Bursztein S, Heifetz M: Cardiac arrest in a child with an anterior mediastinal mass. Anesth Analg 1985; 64:1129-30.
Johnson D, Hurst T, Cujec B, Mayers I: Cardiopulmonary effects of an anterior mediastinal mass in dogs anesthetized with halothane. ANESTHESIOLOGY 1991; 74:725-36.
Fuhrman BP, Everitt J, Lock JE: Cardiopulmonary effects of unilateral airway pressure changes in intact infant lambs. J Appl Physiol 1984; 56:1439-48.
Azarow KS, Pearl RH, Zurcher R, Edwards FH, Cohen AJ: Primary mediastinal masses: A comparison of adult and pediatric populations. J Thorac Cardiovasc Surg 1993; 106:67-72.