Abstract

Resuscitation in a position other then supine is necessary in certain situations. It has been reported that it can be successfully carried out in the prone position but not in the lateral position. The technical limitations and ventilation-perfusion mismatch in the lateral position make the resuscitation difficult. We present a case of successful resuscitation in the lateral position of an infant with a huge anterior mediastinal mass.

Keywords

cardiopulmonary resuscitation; lateral decubitus position; mediastinal neoplasms; posture;

1. Introduction

Anesthesia for a pediatric patient with an anterior mediastinal tumor is a challenge to the anesthesiologist. Severe airway compression and hemodynamic collapse could occur after a change in posture during transportation. We present a case of successful resuscitation in the lateral position of a 4-month-old male infant with a large anterior mediastinal tumor. Hemodynamic collapse occurred after he was changed from the prone to the supine position after being transferred to the operating table from the transportation trolley.

2. Case Report

A 4-month-old male infant weighing 7.5 kg was admitted to our hospital because of tachypnea and vomiting for 1 day. He was born by natural delivery at a gestational age of 40 weeks with birth weight of 3220 g and he had been in good health until this admission. On physical examination, decreased right lung breathing sounds were noted. Chest radiography showed a large anterior mediastinal mass over the right upper chest and collapse of the right lower lobe (Figure 1). Computed tomography (CT) of the chest showed that the mediastinum was displaced to the left posterior aspect by the anterior mediastinal mass, with poor demarcation of the superior vena cava and aortic arch (Figure 2). Because of aggravation of respiratory distress, he was intubated with a conventional pediatric endotracheal tube (inner diameter, 3.5 mm) for ventilator support and was transferred to the pediatric intensive care unit (PICU). Antibiotics were administered to treat the pneumonia mainly on the right lower lobe, and a chest tube was inserted to drain the persistent right pleural effusion. CT-guided tumor biopsy was performed for pathologic diagnosis. During the PICU stay, it was noted that the airway pressure was lower in the prone than in the supine position. However, the peak airway pressure increased progressively from 25 cmH₂O to 38 cmH₂O in 2 days even in the prone position. Arterial CO₂ partial pressure increased to 90 mmHg in spite of adjusting the ventilator settings. Debulking surgery was indicated to relieve the tumor compression.

On the day of surgery, he was transferred to the operating room in the prone position with ventilator support. Blood pressure, heart rate and O₂ saturation were monitored during transportation. On arrival, his complexion was cyanotic with a pulse oximetry that read 85% and the heart rate was around 160 beats/ minute. He was changed from the prone position to the supine position upon transfer to the operating table. After establishment of the monitoring system (electrocardiograph, blood pressure device, pulse oximeter and capnograph), absence of end-tidal CO₂ display was noted. Dislodgment of the endotracheal tube was suspected initially. Bilateral breathing sounds were not audible. The en dotracheal tube was correctly in place, as confirmed by direct laryngoscopy. The ventilator support was set to volume control mode. However, the airway pressure increased to above 50 cmH₂O, and SpO₂ decreased to 65%. A suc tion tube was inserted into the endotracheal tube for removing sputum, but resistance was encountered during the act of insertion. Airway compression by the anterior mediastinal mass was suspected. Therefore, to overcome the compression we intended to push the endotracheal tube deeper. However, breathing sounds and end-tidal CO₂ were still absent. Sinus bradycardia soon developed with a decrease in heart rate to 60 beats/minute and blood pressure by arterial-line became unobtainable. Atropine 0.5 mg was given twice in quick succession but without obvious response. Epinephrine 0.05 mg was given subsequently but it was futile. Arterial blood gas analysis revealed severe respiratory acidosis with pH 6.79, CO₂ partial pressure 313.3 mmHg, O₂ partial pressure 11.5 mmHg, and bicarbonate 48.8 mmol/L. Cardiac massage was performed immediately with both hands to encircle the chest of the infant while leaving the thumbs to apply sternal compression at the junction of the lower and middle third of the sternum. The ventilator was converted to manual bag ging mode during cardiac massage. We performed chest compression at a rate of 100 per minute with manual ventilation at 8−10 breaths/minute. The arterial systolic pressure could be barely maintained at 50−60 mmHg, but high airway pressure and absence of end-tidal CO₂ remained unchanged.

Considering that the mass mainly encroached on the right chest, we contemplated that the conversion of the position from prone to the right lateral could probably ease the airway compression and ventilate the left lung as a result of gravity. Cardiopulmonary resuscitation was then performed in the right lateral position. However, movement of the infant’s body made cardiac compression inefficient. It was not until a hard board was placed at the back of the infant that the cardiac massage could be performed more effectively. With all these maneuvers, bilateral breathing sounds could be heard again, although fainter on the right side. End-tidal CO₂ initially read above 60 mmHg and then decreased to 40 mmHg. SpO₂ gradually increased to 95%. Sinus rhythm resumed and heart rate increased to 140− 150 beats/minute after a 30-minute cardiopulmonary resuscitation. The arterial systolic blood pressure was around 100−110 mmHg. The operation was canceled, and the infant was sent back to the PICU in the right lateral position.

The pathological report of the previous CTguided biopsy revealed a T-cell lymphoblastic lymphoma. He received steroids and chemotherapy, but regression of the tumor was not obvious. Acute respiratory distress syndrome, pneumothorax, and pneumomediastinum set in 3 days later. In spite of application of extracorporeal membrane oxygenation, refractory septic shock developed, and the infant died 2 weeks later.

Download full-size image

Figure 1 Chest radiograph shows a large anterior medi- astinal mass over the right upper chest and collapse of the right lower lung lobe.

Download full-size image

Figure 2 Chest computed tomography reveals a large anterior mediastinal mass (white arrow) compressing the mediastinum to the left posterior side (black arrow = heart; white star = collapsed right lung; black star = aorta).

3. Discussion

We present a case of successful resuscitation carried out in the right lateral position in an infant with a large anterior mediastinal tumor. It reminded us that cardiopulmonary resuscitation in a position other than supine is necessary in some specific circumstances.^1,2 There are some reports concerning resuscitation in the prone position,^3,4 yet no resuscitation carried out in the lateral position has ever been reported. For more effective resuscitation in the lateral position, it is essential that another resuscitator should hold a hard board against the back of the patient. For an infant, the standard method for the supine position could likewise be performed in the lateral position. With the fingers of both hands encircling the chest while leaving both thumbs to perform compression at the junction of the lower and middle third of the sternum, cardiopulmonary resuscitation could be carried out. Effective cardiac output is mainly determined by adequate stroke volume and regular heart beats. Because cardiac output in children is more dependent on heart rate than in adults, bradycardia requires immediate intervention.

An anterior mediastinal tumor may pose exceptional risks in anesthesia.^5,6 First, in the supine position, the decreased volume of the lungs and the gravitational force of the tumor may partially or completely compress the lumen of the airway.⁷⁻⁹ Second, the tumor may press down further after the muscle tone is weakened by anesthesia and may cause further airway compression. Even if the patient can tolerate the ordeal in the supine position before anesthesia, severe respiratory compromise can occur after deep anesthesia and/or neuromuscular blockade by a muscle relaxant.¹⁰ Third, the anterior mediastinal tumor could also compress the pulmonary artery in the supine position, leading to hypoxemia, hypotension, or even cardiac arrest.^6,11 Therefore, maintaining spontaneous ventilation is mandatory as suggested by most reports in the literature.^7,12,13 Detailed evaluation of airway patency and hemodynamic stability and careful anesthetic planning are of paramount importance. Preoperative anesthetic planning should include the position of the patient during transportation and discussion with the surgeons about the preparation for cardiopulmonary bypass and the feasibility of doing immediate median sternotomy should the tumor compression cause serious hemodynamic and pulmonary problems requiring corrections beyond the scope of ordinary measures.

There are many reasons to explain why resuscitation in the lateral position was vital in this infant. First, the anterior mediastinal tumor in this infant encroached upon most of the right chest cavity, and thus compressed both the trachea and bronchi. Second, the functional condition was worse in the right lung than in the left lung because of collapse of the right lower lobe with pleural effusion. Therefore, placing the patient in the lateral position could relieve the compression on the left bronchus to ease the working of the left lung. There might be a disadvantage in the right decubitus position, because the mediastinal mass might compress the superior vena cava leading to superior vena cava syndrome. We must be aware that drugs may be given in vain in the upper limbs. If superior vena cava syndrome does occur,^8,14−16 resuscitation in the prone position may be an alternative way. The effects of gravity may draw the anterior mediastinal mass downward to ease the compression on the airway and relieve the hindrance to the cardiac output. Reverse precordial compression might be performed with the fingers of both hands encircling the chest, while leaving both thumbs to do compression on the back. Moreover, the lateral position would aggravate ventilation-perfusion mismatch. However, in our case, the huge anterior mediastinal mass might compress the right pulmonary artery in the right lateral body position, and thus it could reduce shunting in the dependent lung. When manpower and equipment are available, rigid bronchoscopy may be a life-saving measure to relieve the airway compression. Furthermore, median sternotomy with cardiopulmonary bypass can be life-saving.

In this case, pneumothorax and pneumomediastinum developed 3 days after resuscitation. Pulmonary air leakage in small infants is often related to iatrogenicity such as cardiopulmonary resuscitation and mechanical ventilation. In our case, pulmonary barotrauma during cardiopulmonary resuscitation was less likely because the pneumothorax and pneumomediastinum appeared 3 days later. Subsequent mechanical ventilation might be the cause of pulmonary barotrauma. Because the airway was compressed by the anterior mediastinal tumor, higher airway pressure was required to achieve adequate tidal volume. In addition, a patient with acute respiratory distress syndrome usually has poor pulmonary compliance, which could render the patient more susceptible to pulmonary barotrauma.

In conclusion, anesthesia for patients with an anterior mediastinal tumor requires detailed evaluation of preoperative airway patency and hemodynamic stability. The risks and benefits of biopsy and operation should be seriously balanced. We suggest that for a pediatric patient with unilateral anterior mediastinal tumor, cardiopulmonary resuscitation can be carried out in the lateral position, provided that a hard board is placed at the patient’s back. For adults, resuscitation in the lateral position may not be as effective as in infants, and further investigation is needed.