TENSION pneumothorax is a life-threatening emergency that rapidly results in cardiopulmonary arrest. Although often associated with trauma or central line placement, tension pneumothorax can also occur intraoperatively from positive pressure ventilation or from the surgical procedure. Although some case reports have described tension pneumothorax during video-assisted thoracoscopic surgery, 1,2we report a tension pneumothorax that developed while the chest was open following median sternotomy and entrance into the contralateral pleural space.

A 66-yr-old man (84.5 kg, 170 cm) with severe chronic obstructive pulmonary disease (COPD) presented for lung volume reduction surgery (LVRS). His past medical history was significant for tobacco abuse and severe COPD. On physical examination, the patient had conversational dyspnea, clubbing, peripheral cyanosis, and bilateral expiratory wheezing in all lung fields. A preoperative chest radiograph showed a normal-sized heart, increased anteroposterior diameter, and flattening of the hemidiaphragms. A computed tomography scan showed evidence of severe emphysema with bilateral upper lobe predominance. The bullae were the target lesions for the surgical procedure. 3Preoperative pulmonary function test results are shown in table 1. The patient's arterial blood gas data while breathing room air were as follows:p  H, 7.40; Paco2, 48 mmHg; Pao2, 63 mmHg; bicarbonate, 29 mEq/l; and arterial oxygen saturation, 93%. Right heart catheterization (table 2) indicated elevated pulmonary artery pressures.

Table 1. Preoperative Pulmonary Function Test Results

Because the patient had emphysematous target lesions on chest computed tomography scan, a carbon monoxide diffusing capacity was not performed.

ERV = expiratory reserve volume; FEF25–75= midexpiratory (25–75%) forced expiratory flow; FEV1= forced expiratory volume in 1 s; FVC = forced vital capacity; MVV = maximum voluntary ventilation; RV = residual volume; TLC = total lung capacity.

Table 1. Preoperative Pulmonary Function Test Results
Table 1. Preoperative Pulmonary Function Test Results

Table 2. Preoperative Right Heart Cardiac Catheterization Results

Additional preoperative cardiac workup was unremarkable. Electrocardiogram demonstrated a normal sinus rhythm and low voltage QRS complexes consistent with pulmonary disease. A dobutamine–thallium scan showed no scintigraphic evidence of myocardial ischemia or prior myocardial infarction. Transthoracic echocardiography showed no significant valvular disease and a left ventricular ejection fraction of 60%.

Table 2. Preoperative Right Heart Cardiac Catheterization Results
Table 2. Preoperative Right Heart Cardiac Catheterization Results

A thoracic epidural catheter was placed at the T6 level using a right paramedian hanging drop technique without difficulty and was dosed with 10 ml ropivacaine, 0.5%, in divided doses. The patient had an uneventful induction, and maintenance anesthesia was achieved with sevoflurane in oxygen. A 39-French, left-sided, double-lumen endotracheal tube was placed without difficulty, and positioning was confirmed via  fiberoptic bronchoscopy. Ventilation was provided via  pressure control mode (peak pressure = 30 cm H2O) at a rate of 14 breaths/min. Delivered tidal volumes during left one-lung ventilation were approximately 375 ml, with a set inspiratory to expiratory ratio of 1:3.5. A central venous catheter was placed without complication via  the right subclavian vein.

A midline sternotomy was performed, and the surgeon opened the right lung pleura. Approximately 25 min after initiating right pulmonary resection, there was a sudden, significant loss of tidal volume and rapid (over several seconds) development of pulseless electrical activity. The patient exhibited profound hypotension (nonpulsatile arterial line with pressures < 30 mmHg), a sinus rhythm without ST segment changes, and decreased end-tidal carbon dioxide. Initial treatment consisted of open cardiac massage and a bolus dose of epinephrine (250 μg) while determining the cause of the hypotension. Inspection of the operative field revealed that the left pleura tented (under pressure) past the midline. Opening of the left pleura was associated with a rush of air and resolution of the hypotension within seconds. Following an initial transient sinus tachycardia and hypertension secondary to the epinephrine, the vital signs returned to baseline, and surgery continued. There was a continual large air leak from the left lung. A ruptured bulla of the left upper lobe was noted and excised with other diseased lung tissue. The patient was extubated at the conclusion of the procedure. A persistent bronchopleural fistula of the left upper lobe required reoperation 3 weeks after the initial procedure. The remainder of the perioperative course was unremarkable.

A tension pneumothorax may occur when intrapleural pressure exceeds atmospheric pressure. This usually occurs from pleural injury in which a conduit acts like a one-way valve, allowing air to escape into the pleural space. Expansion of the pneumothorax forces the lung to collapse, with resultant increasing airway pressures. Increasing intrathoracic pressure exceeds venous and cardiac end-diastolic pressures, limiting cardiac filling, stroke volume, and blood pressure (pulseless electrical activity). Tension pneumothorax is a medical emergency and requires immediate decompression. Although case reports of tension pneumothorax exist in the setting of thoracotomy and video-assisted thoracoscopic surgery, 1,2there are no published case reports of tension pneumothorax in a patient with an open chest following sternotomy.

The differential diagnosis in our case includes pulmonary thromboembolism, venous air embolism, torsion of the heart or great vessels, compression of the heart by the surgeon, primary cardiac etiologies (dysrhythmia, myocardial ischemia), “pulmonary tamponade” resultant from air trapping, acute right heart failure secondary to marked pulmonary hypertension, malposition of the double-lumen endotracheal tube, and iatrogenic pneumothorax from the thoracic epidural or central venous cannulation procedures. The rapid resolution of the hypotension and the rush of air associated with opening of the left pleura confirm the diagnosis of tension pneumothorax. Iatrogenic causes do not seem likely. The central line was placed on the contraleteral side. Bilateral dermatomal anesthesia, low volatile agent requirements intraoperatively, and excellent postoperative analgesia suggest placement of the epidural catheter into the epidural space rather than the pleural space. Malposition of the double-lumen endotracheal tube can result in barotrauma. Fiberoptic bronchoscopy was performed before and after the development of the pneumothorax and demonstrated proper positioning of the endotracheal tube. The most likely cause of the pneumothorax was rupture of a pulmonary bulla secondary to positive pressure ventilation. Because the left pleura was not yet opened, a tension pneumothorax ensued.

The LVRS population is selected for severe obstructive disease and is therefore at high risk for intraoperative pneumothorax. LVRS has had limited success as a palliative procedure for severe COPD, with 30-day mortality rates ranging from 0 to 15%. 4–6Pneumothorax-related complications accounted for approximately 3% of the American Society of Anesthesiologists’ closed claims database. 7Barotrauma was responsible for 20% of the pneumothorax cases, with placement of brachial plexus blocks and central lines comprising the majority of claims. 7Pneumothorax may occur in 3–5% of mechanically ventilated patients and is life-threatening if a tension pneumothorax develops. Patients with COPD have an increased risk of pneumothorax. 8Previous reports of LVRS via  sternotomy did not report intraoperative pneumothorax but did advocate ventilator management of pressure control ventilation (20–25 cm H2O) with a prolonged inspiratory to expiratory ratio of 1:4, 4for example, which was similar to our management. The prolonged inspiratory to expiratory ratio is critically important, as air trapping can result in a physiologic process that resembles tension pneumothorax.

Early postoperative extubation in LVRS patients decreases work of breathing and airway resistance and increases dynamic compliance. 9Extubation also protects the lungs from prolonged barotrauma and its associated risks, such as persistent air leaks and pneumothorax. Although the majority of LVRS patients are extubated in the early postoperative period, approximately 6–10% require reintubation. This group of patients has significantly higher morbidity and mortality rates. 10Because the patient described here met our extubation criteria, he was extubated without complication. Many thoracic anesthesiologists feel that a thoracic epidural block is the most appropriate method of achieving postoperative analgesia. This is relevant because thoracic epidural catheter placement can result in pneumothorax if the needle is off midline and enters the pleural space. 11 

This is a case of tension pneumothorax during median sternotomy. A priori , this would seem impossible with an open chest. However, the human anatomy is characterized by two separate pleural cavities. Anesthesiologists routinely capitalize on this anatomic arrangement when using one-lung ventilation. One-lung ventilation would not be technically feasible without the use of positive end-expiratory pressure on the dependent lung if the pleura were contiguous. Furthermore, the pericardium was intact; thus, the heart and great vessels were exposed to the pressure effects of the two pleural spaces rather than atmospheric pressure. We believe that the mechanism was compression and displacement of the heart, leading to decreased venous return. Normal hemodynamic parameters were immediately restored following release of the left-sided pneumothorax. The lesson to be learned from this case is that whenever the pleural cavity is intact, conditions for the development of a tension pneumothorax exist and must be considered in the event of hemodynamic collapse.

Zollinger A, Zaugg M, Weder W, Russi EW, Blumenthal S, Zalunardo MP, Stoehr S, Thurnheer R, Stammberger U, Spahn DR, Pasch T: Video-assisted thoracoscopic volume reduction surgery in patients with diffuse pulmonary emphysema: Gas exchange and anesthesiological management. Anesth Analg 1997; 84: 845–51
Roush TF, Crawford AH, Berlin RE, Wolf RK: Tension pneumothorax as a complication of video-assisted thoracoscopic surgery for anterior correction of idiopathic scoliosis in an adolescent female. Spine 2001; 26: 448–50
National Emphysema Treatment Trial Research Group: Patients at high risk of death after lung-volume-reduction surgery. N Engl J Med 2001; 345: 1075–83
National Emphysema Treatment Trial Research Group:
Buettner AU, McRae R, Myles PS, Snell GI, Bujor MA, Silvers A, Weeks AM: Anaesthesia and postoperative pain management for bilateral lung volume reduction surgery. Anaesth Intensive Care 1999; 27: 503–8
Flaherty KR, Kazerooni EA, Curtis JL, Iannettoni M, Lange L, Schork MA, Martinez FJ: Short-term and long-term outcomes after bilateral lung volume reduction surgery: Prediction by quantitative CT. Chest 2001; 119: 1337–46
Flaherty KR, Martinez FJ: Lung volume reduction surgery for emphysema. Clin Chest Med 2000; 21: 819–48
Cheney FW, Posner KL, Caplan RA: Adverse respiratory events infrequently leading to malpractice suits. A closed claims analysis. A nesthesiology 1991; 75: 932–9
Jantz MA, Pierson DJ: Pneumothorax and barotrauma. Clin Chest Med 1994; 15: 75–91
Tschernko EM, Gruber EM, Jaksch P, Jandrasits O, Jantsch U, Brack T, Lahrmann H, Klepetko W, Wanke T: Ventilatory mechanics and gas exchange during exercise before and after lung volume reduction surgery. Am J Respir Crit Care Med 1998; 158: 1424–31
Keller CA, Naunheim KS: Perioperative management of lung volume reduction patients. Clin Chest Med 1997; 18: 285–300
Zaugg M, Stoehr S, Weder W, Zollinger A: Accidental pleural puncture by a thoracic epidural catheter. Anaesthesia 1998; 53: 69–71