Secondary Logo

Journal Logo

Life-Threatening Hypoxemic Respiratory Failure After Repair of Acute Type A Aortic Dissection: Successful Treatment with Venoarterial Extracorporeal Life Support Using a Prosthetic Graft Attached to the Right Axillary Artery

Yokota, Kimio MD; Fujii, Tomoko MD; Kimura, Kenichi MD; Toriumi, Takashi MD; Sari, Atsuo MD

doi: 10.1097/00000539-200104000-00014

Department of Anesthesiology and Intensive Care Medicine, Kawasaki Medical School, Kurashiki, Okayama, Japan

December 14, 2000.

Address correspondence and reprint requests to Kimio Yokota, MD, Department of Anesthesiology and Intensive Care Medicine, Kawasaki Medical School 577. Matsushima, Kurashiki-City, Okayama 701–0192, Japan.

IMPLICATIONS: We report the successful treatment of life-threatening hypoxemic respiratory failure with venoarterial extracorporeal life support (VA ECLS) using a prosthetic graft attached to the right axillary artery after repair of acute type A aortic dissection.

Deterioration in pulmonary gas exchange occurs after cardiovascular surgery involving cardiopulmonary bypass (CPB) (1–6). A cohort study reported a 12% incidence and a 4.4% mortality in 1,461 patients with early onset acute pulmonary dysfunction after CPB, which was defined on the patient’s arrival at the intensive care unit (ICU) and the need for mechanical ventilation with a Pao2/inspired oxygen fraction (Fio2) ratio of <150 mm Hg (1). We report the successful treatment of life-threatening hypoxemic respiratory failure with venoarterial extracorporeal life support (VA ECLS) using a prosthetic graft attached to the right axillary artery after repair of acute type A aortic dissection.

Back to Top | Article Outline

Case Report

A 72-yr-old woman (approximate height 155 cm and weight 50 kg) suddenly complained of severe retrosternal pain after defecation. She had a history of smoking and hypertension but no history of pulmonary disease. A supine portable chest film showed no abnormal pulmonary parenchymal shadows, a cardiothoracic ration (CTR) of 0.65 and widening of the aorta. Thoracic computed tomography (CT) with contrast confirmed an acute dissection of the ascending and arch aorta and periocardial effusion. The patient was emergently transferred to the operating theater. Subsequent to the application of standard monitoring devices, the patient’s heart rate (HR) decreased rapidly to 40 bpm, and the electrocardiograph monitor showed a widening of the QRS complex and increase of the ST segment. The radial arteries were not palpable. The patient became unresponsive. Anesthesia was immediately induced with 50 μg fentanyl, 2.5 mg midazolam, 10 mg vecuronium and 100% oxygen. After a tracheal tube (inner diameter, 8.0 mm) was placed, the HR increased to 120 bpm. The electrocardiograph monitor showed a narrowing of the QRS complex and improvement in the ST segment. Catheters (20-gauge) were percutaneously inserted in the left and right brachial arteries (BA). The patient’s blood pressure (BP) was 70/50 mm Hg. Arterial blood gas analysis (ABG) showed pH 6.91, Paco2 48 mm Hg, Pao2 138 mm Hg, and a base deficit of 20 mEq/L while breathing 100% O2. Continuous infusion of dopamine (10 μg · kg−1 · min−1) was initiated to treat systemic hypotension. A triple-lumen central venous catheter with a large bore size (12F) was percutaneously inserted through the right internal jugular vein. A catheter (22-gauge) was surgically inserted in the left superficial temporal artery. Rectal temperature was continuously measured. Anesthesia was maintained by small doses of midazolam and fentanyl.

After administration of 8000 U heparin, activated coagulation time was prolonged to 937 s. A cannula (18F) was inserted into the right femoral artery and another cannula (19F/50 cm) was inserted into the inferior vena cava through the right femoral vein. Partial CPB was initiated at a flow rate of 2.0 L/min. The membrane oxygenator delivered an oxygen concentration of 60%. The ABG measured from the left BA showed pH 7.06, Paco2 52 mm Hg, Pao2 129 mm Hg, and a base deficit of 13 mEq/L while breathing 100% O2. After median sternotomy, a large single venous cannula (36F) was inserted into the right atrium, and total CPB was established. The patient was cooled to 18°C measured by the rectal temperature. An aortic cross-clamp was placed just proximal to the brachiocephalic artery. The ascending aorta was opened, and crystalloid cardioplegia was directly infused into the ostia of the coronary arteries. The aortic arch was opened, and the balloon catheters (MD-25315; Sumitomo Bakelite, Co. Ltd., Tokyo, Japan) were individually inserted into the brachiocephalic (15F), left carotid (15F), and left subclavian arteries (12F). Then the right femoral arterial inflow rate was reduced to 200 mL/min and cerebral protection was achieved by antegrade selective cerebral perfusion of a flow rate of 850 mL/min. The left superficial temporal artery pressure was maintained at a range of 40–60 mm Hg. The dissection originated from the ascending aorta just distal to the coronary orifices and extended to the descending aorta just distal to the opposite wall of the left subclavian artery origin. The dissection involved approximately 50% of the aortic arch circumference and did not involve the origins of the branch vessels of the aortic arch. Therefore, ascending aorta replacement with hemiarch replacement was performed. A 30 mm Dacron® graft was anastomosed to the aortic arch without the use of a distal cross-clamp. After distal graft anastomosis, the graft was cross-clamped, the right femoral arterial inflow rate was increased to 2.0 L/min, and selective cerebral perfusion was discontinued. The time of decreased femoral arterial inflow rate was 49 min. The patient was rewarmed. The surgeons discovered injury to the brachiocephalic artery bifurcation resulting from insertion of the balloon catheter and proceeded to repair the injury completely. During the repair, rewarming was interrupted and the rectal temperature was maintained at approximately 20°C to protect the brain because intermittent cross-clamping of the brachiocephalic, right common carotid, and right subclavian arteries was required.

After the completion of proximal graft anastomosis, the graft cross-clamp was removed. The aortic cross-clamp time was 195 min. Defibrillation was successfully accomplished using internal paddles with an energy of 30 J. Mechanical ventilation was restarted. The patient was gradually weaned from CPB. However, she bled excessively, so multiple units of homologous bloods were transfused to maintain mean arterial pressure at more than 40 mm Hg. The surgeons discovered that arterial blood originating from the suture lines was desaturated, although mechanical ventilation was working optimally. Pulse oximetry showed that peripheral arterial hemoglobin oxygen saturation (Spo2) was 85%. Subsequent to the cessation of CPB, HR decreased to 30 bpm and CPB was therefore reinstituted. The ABG obtained from the left BA during separation from CPB showed pH 7.36, Paco2 36 mm Hg, and Pao2 48 mm Hg while breathing 100% O2. After restoring the femoral arterial inflow rate to 2.0 L/min, the HR increased to 120 bpm, but the Pao2 remained unchanged at 51 mm Hg while breathing 100% O2.

The working diagnosis was severe hypoxemic respiratory failure. After the femoral arterial inflow rate was increased to 3.0 L/min, a horizontal incision was made below the medial third of the right clavicle, and an 8-mm woven Dacron® graft (Meadox Medicals, Inc, Oakland, NJ) was anastomosed end-to-side to the right axillary artery. A flexible arterial cannula (19F) was introduced and advanced approximately 10 cm into the prosthetic graft. After establishment of VA ECLS with venous drainage from the inferior vena cava and arterial inflow into the prosthetic graft, CPB was successfully discontinued. The CPB time was 364 min. A flow rate of 2.0 L/min was maintained during VA ECLS using a centrifugal pump (Bio Pump® BP-80; Medtronic Inc., Eden Prairie, MN). The membrane oxygenator (Menox® AL-6000; Kuraray Co. Ltd., Kurashiki, Japan) delivered an oxygen concentration of 100%. The ABG obtained from the left BA showed pH 7.19, Paco2 33 mm Hg, and Pao2 284 mm Hg while breathing 100% O2. Heparin-induced anticoagulation was reversed with 64 mg prolamine.

Although anatomical disruption warranting surgical intervention was not evident, extensive bleeding continued. The blood salvage system was used. After administration of an additional 32 mg prolamine, activated coagulation time remained prolonged at 210 s. Because the platelet and plasma replacement therapy did not reduce the diffuse bleeding, surgeons requested that the patient’s family ask volunteers with a matching blood type to donate fresh whole blood. The coagulation abnormality dramatically improved in response to the transfusion of 2000 mL of irradiated fresh whole blood collected from five volunteers. The patient’s estimated blood loss was 25000 mL. Crystalloid solutions of 8200 mL and colloid solutions of 3500 mL were infused. Packed red blood cells of 80 U, fresh-frozen plasma of 40 U, platelet concentrates of 10 U, and salvaged blood of 2800 mL were transfused. Total urine output was 5000 mL. As the patient hemodynamically stabilized with inotropic drug support (10 μg · kg−1 · min−1 dopamine and 0.5 μg · kg−1 · min−1 milrinone), drainage tubes were inserted into the right pleural and pericardial spaces, and the chest was closed.

In the ICU, the patient had a BP of 130/90 mm Hg and HR of 115 bpm. Her hematocrit was 45%. The activated coagulation time returned to 139 s. The chest film revealed slight pulmonary vascular congestion, cardiac enlargement (CTR of 0.69), left pleural effusion, and bilateral elevation of the diaphragm. Postoperative mechanical ventilation was achieved using a Servo 300® ventilator (Siemens-Elema, Solma, Sweden) using pressure-controlled ventilation as the ventilation mode. Ventilation variables were initially an Fio2 of 1.0, a set pressure (Pset) of 12 cm H2O, a respiratory rate of 12 breaths/min, and positive end-expiratory pressure (PEEP) of 5 cm H2O. The ABG from the left BA showed pH 7.50, Paco2 30 mm Hg, and Pao2 128 mm Hg. The Pset was carefully adjusted to maintain the tidal volume at <6 mL/kg during VA ECLS. Total bleeding into the pleural and pericardial spaces occurred at a rate of 50–100 mL/h. Heparin was titrated to achieve an activated coagulation time of 160 to 180 s during VA ECLS.

Bleeding discontinued on the second postoperative day and the patient had completely emerged from anesthesia at that time. Neurological deficit was absent. On the third postoperative day, mechanical ventilation was set to deliver an Fio2 of 0.9, a Pset of 16 cm H2O, a respiratory rate of 12 breaths/min, and PEEP of 8 cm H2O. Blood gas analyses of the right brachial, right superficial temporal, left superficial temporal, left brachial, and left femoral arteries were analyzed during pump flow of 2.0 L/min and 90% O2 delivery using a membrane oxygenator (Table 1). On the fifth postoperative day, a drainage tube was inserted into the left pleural space because of an increase in left pleural effusion.

Table 1

Table 1

On the sixth postoperative day, a chest radiograph revealed clear lung fields and a decreased cardiac size (CTR of 0.53). The ABG from the left BA (pH 7.46, Paco2 34 mm Hg, and Pao2 129 mm Hg while breathing 70% O2) suggested that pulmonary oxygenation was compatible with cessation of VA ECLS. The patient’s hemodynamics were stabilized with 6 μg · k−1 · min−1 dopamine and 0.5 μg · k−1 · min−1 milrinone. Separation from VA ECLS was initiated with pressure-controlled ventilation mode (Fio2 of 0.9, respiratory rate of 12 breaths/min, Pset of 12 cm H2O, and PEEP of 8 cm H2O) while Spo2 was carefully monitored. After cessation of VA ECLS, pulse oximetry showed Spo2 of 100% and the patient’s hemodynamic status remained stable. Mechanical ventilation was continued without change. Approximately 10 min after separation from VA ECLS, Spo2 remained unchanged at 100%, and Fio2 was gradually reduced to 0.6. The ABG showed pH 7.39, Paco2 40 mm Hg, and Pao2 98 mm Hg. The prosthetic graft attached to the right axillary artery was surgically removed. After removal of the prosthetic graft, neither brachial plexus palsy nor axillary artery thrombosis was evident.

On the tenth postoperative day, the tracheal tube was removed, because pulmonary oxygenation had improved (Pao2 87 mm Hg on Fio2 of 0.4). However, upper airway obstruction persisted despite appropriate head positioning. After confirmation of bilateral vocal cord palsy using a fiberscope, the patient was again intubated. A tracheostomy was performed on the 18th postoperative day. On the 31st postoperative day, the patient was separated from mechanical ventilation and discharged from the ICU. On the 43rd postoperative day, the otolaryngologist confirmed an improvement in vocal cord palsy. On the 59th postoperative day, the tracheostomy tube was removed, and the patient was discharged on the 71st postoperative day.

Back to Top | Article Outline


The causes of pulmonary dysfunction after CPB are multifactorial (2). Increased extravascular lung water promoted by hypooncotic bypass prime solution and lung damage associated with surgical procedures during periods of heparinization may be partially responsible for hypoxemia after CPB. An experimental study using thoracic CT demonstrated that severe atelectasis in the dorsal dependent lung regions, which is unlikely to be detected by conventional supine anteroposterior chest radiography, plays an important role in hypoxemia after CPB (3). However, a clinical study reported that pulmonary shunt does not correlate with atelectasis quantified by thoracic CT on the first postoperative day in coronary artery bypass graft patients (4). Recent reports indicate that the systemic inflammatory response to CPB may be associated with pulmonary dysfunction after CPB (5,6). During surgical procedures using CPB, blood contact activation, ischemia and reperfusion injury, and endotoxemia contribute concurrently to set off a cascade of proinflammatory events (7). Stimulated neutrophils are retained in the pulmonary capillaries as a result of a decreased ability of the cell to deform (8), and adhesion of these neutrophils to the endothelial surface might enhance this retention (9). Stimulated neutrophils release intracellular lysosomal granular contents, such as elastase, myeloperoxidase and oxygen-derived free radicals. Consequently, neutrophils damage the pulmonary endothelium, thus increasing pulmonary capillary permeability. However, the exact pathophysiologic mechanisms for CPB-associated acute lung injury remain inadequately defined.

Profound shock impairs pulmonary function. Animal studies show that hypovolemic shock causes pulmonary ultrastructural morphologic abnormalities within one hour of onset (10). Pathologic findings include endothelial and interstitial edema and significant cellular infiltration of lung tissue, particularly by leukocytes. Therefore, hypotension and bradycardia before the induction of anesthesia may have been related to the development of hypoxemic respiratory failure after CPB in the present case. In fact, pulmonary oxygenation was already impaired before the institution of CPB. Conversely, the multiple homologous blood transfusions may have been responsible for the hypoxemic respiratory failure after CPB because transfusion-related acute lung injury occurs within four hours after transfusion and is characterized by hypoxemia resulting from noncardiogenic pulmonary edema (11). Transfusion-related acute lung injury probably results from several mechanisms, including the reaction of blood-donor antibodies that have neutrophil antigenic specificity with the recipient’s neutrophils, leading to increased pulmonary capillary permeability (11). In addition, one study showed that the transfusion of biologically active lipids in stored blood, which are capable of neutrophil priming, is associated with transfusion-related acute lung injury (12).

Extracorporeal membrane oxygenation (ECMO) provides effective treatment for adult patients who suffer severe hypoxemic respiratory failure (13,14). The rapid type of ECMO entry criteria includes Pao2 of <50 mm Hg on an Fio2 of 1.0 and Paco2 of <45 mm Hg (15). Our patient satisfied the ECMO entry criteria based on pulmonary gas exchange. Two recent studies reported that survival rate with ECMO therapy was approximately 55% for adults (13,14). Several clinical disorders are associated with the development of acute hypoxemic respiratory failure. Kolla et al. (14) reported that 2 of 94 patients with hypoxemic respiratory failure receiving ECMO therapy underwent cardiac surgery and one of these patients survived. The results of the study of Kolla et al. (14) indicate that life-threatening hypoxemic respiratory failure occurs less commonly after CPB. There have been no clinical reports of the survival rate for ECMO therapy in adults with life-threatening hypoxemic respiratory failure after CPB.

The axillary artery provides an excellent arterial access for CPB (16–18). The arterial inflow method through the axillary artery can involve the direct cannulation or the graft interposition techniques. In the present case, we used the prosthetic graft attached to the right axillary artery as the arterial inflow site for VA ECLS, although this necessitated additional surgery at a later date. VA ECLS involves more prolonged arterial access than CPB. Although the axillary artery enjoys rich collateralization, prolonged direct cannulation of the axillary artery may cause ischemia of the upper extremity, particularly when the arterial cannula completely occludes the axillary artery. Furthermore, the axillary artery is often fragile, and the inflow cannula may cause trauma to the artery. However, the graft interposition technique maintains perfusion to the upper extremity and avoids any contact between the friable artery and the inflow cannula. Sabik et al. (16) reported the association of the direct cannulation technique with local complications including a mild reversible brachial plexus palsy and axillary artery thrombosis. The brachial plexus roots are located lateral to the axillary artery. A medial approach to the axillary artery is better facilitated by the graft interposition technique than by the direct cannulation method (18).

In the current case, VA ECLS using antegrade perfusion through the graft attached to the right axillary artery supplied oxygenated blood to the left subclavian artery intraoperatively, although the left ventricle pumped deoxygenated blood. An experimental study in baboons that induced hypoxemia with an Fio2 of 0.08 has shown that right axillary artery perfusion was more effective in the delivery of well-oxygenated blood to the aortic arch than was femoral artery perfusion during venoarterial bypass (19). On the third postoperative day, we confirmed that arterial inflow through the prosthetic graft delivered adequately saturated blood into the aortic arch, although arterial oxygenation of the distal aortic arch was decreased in comparison with that of the proximal aortic arch.

ECMO only allows injured lungs the chance to heal, and is not a therapy to treat diseased lungs. Most recently, a multicenter prospective randomized controlled clinical trial demonstrated that mortality is significantly decreased in patients with acute respiratory distress syndrome (ARDS) treated with lower tidal volumes (6 mL/kg of predicted body weight) than in those treated with traditional tidal volumes (12 mL/kg of predicted body weight) (20). Therefore, our patient required a protective-ventilation strategy with an Fio2 of <0.4, a tidal volume of <6 mL/kg, and a peak static airway pressure of <30 cm H2O during ECMO. Although we carefully maintained the tidal volume at <6 mL/kg and the peak static airway pressure at <30 cm H2O with pressure-controlled ventilation mode, we did not apply an Fio2 of <0.4 during ECMO. The coronary arteries are perfused mainly with desaturated blood from the left ventricle during ECMO using the right common carotid artery cannulation for the diseased lung (21). Hence, the application of an Fio2 of <0.4 may impair cardiac function during ECMO that is instituted to manage hypoxemic respiratory failure after CPB.

In ARDS patients, the dorsal dependent alveoli are underinflated during mechanical ventilation (22). Excessive stress at margins between aerated and atelectatic regions of the dorsal dependent alveoli may cause lung injury independently of alveolar overdistension (23). In addition, the cyclic collapse and reopening of alveoli could induce both the pulmonary and the systemic cytokine response (24). Based on measurements of the lower inflection point on a pressure-volume curve of ARDS patients, PEEP levels of more than 15 cm H2O are required to keep the lungs open at a tidal volume of 6 mL/kg (25). The “open-lung” approach to mechanical ventilation probably reduces ventilator-associated lung injury. From a respiratory protection standpoint, we believe in hindsight that PEEP of more than 15 cm H2O during VA ECLS may have been beneficial.

In summary, although additional surgery was required, the case presented herein shows that VA ECLS using a prosthetic graft attached to the right axillary artery can provide effective treatment for patients with life-threatening hypoxemic respiratory failure after CPB.

Back to Top | Article Outline


1. Rady MY, Ryan T, Starr NJ. Early onset of acute pulmonary dysfunction after cardiovascular surgery: Risk factors and clinical outcome. Crit Care Med 1997; 25: 1831–9.
2. Matuschak GM. Pulmonary dysfunction after surgery involving cardiopulmonary bypass: do we understand the mechanisms? Crit Care Med 1997; 25: 1778–80.
3. Magnusson L, Zemgulis V, Wicky S, et al. Atelectasis is a major cause of hypoxemia and shunt after cardiopulmonary bypass: an experimental study. Anesthesiology 1997; 87: 1153–63.
4. Tenling A, Hachenberg T, Tydén H, et al. Atelectasis and gas exchange after cardiac surgery. Anesthesiology 1998; 89: 371–8.
5. Kotani N, Hashimoto H, Sessler DI, et al. Neutrophil number and interleukin-8 and elastase concentrations in bronchoalveolar lavage fluid correlate with decreased arterial oxygenation after cardiopulmonary bypass. Anesth Analg 2000; 90: 1046–51.
6. Hensel M, Volk T, Döcke, et al. Hyperprocalcitonemia in patients with noninfectious SIRS and pulmonary dysfunction associated with cardiopulmonary bypass. Anesthesiology 1998; 89: 93–104.
7. Hall RI, Smith MS, Rocker G. The systemic inflammatory response to cardiopulmonary bypass: pathophysiological, therapeutic, and pharmacological considerations. Anesth Analg 1997; 85: 766–82.
8. Worthen GS, Schwab B, Elson EL, Downey GP. Mechanics of stimulated neutrophils: Cell stiffening induces retention in capillaries. Science 1989; 245: 183–6.
9. Kubo H, Doyle NA, Graham L, et al. L-and P- selectin and CD11/CD18 in intracapillary neutrophil sequestration in rabbit lungs. Am J Respir Crit Care Med 1999; 159: 267–74.
10. Pretorius JP, Schlag G, Redi H, et al. The ‘lung in shock’ as a result of hypovolemic-traumatic shock in baboons. J Trauma 1987; 27: 1344–53.
11. Goodnough LT, Brecher ME, Kanter MH, et al. Transfusion medicine, first of two parts: blood transfusion. N Engl J Med 1999; 340: 438–47.
12. Silliman CC, Paterson AJ, Dickey WO, et al. The association of biologically active lipids with the development of transfusion-related acute lung injury: a retrospective study. Transfusion 1997; 37: 719–26.
13. Bartlett RH, Roloff DW, Custer JR, et al. Extracorporeal life support: the University of Michigan experience. JAMA 2000; 283: 904–8.
14. Kolla S, Awad SS, Rich PB, et al. Extracorporeal life support for 100 adult patients with severe respiratory failure. Ann Surg 1997; 226: 544–66.
15. Zapol WM, Snider MT, Hill JD, et al. Extracorporeal membrane oxygenation in severe acute respiratory failure: a randomized prospective study. JAMA 1979; 242: 2193–6.
16. Sabik JF, Lytle BW, McCarthy PM, et al. Axillary artery: an alternative site of arterial cannulation for patients with extensive aortic and peripheral vascular disease. J Thorac Cardiovasc Surg 1995; 109: 885–91.
17. Neri E, Massetti M, Capannini G, et al. Axillary artery cannulation in type A aortic dissection operations. J Thorac Cardiovasc Surg 1999; 118: 324–9.
18. Baribeau YR, Westbrook BM, Charlesworth DC, et al. Arterial inflow via an axillary artery graft for the severely atheromatous aorta. Ann Thorac Surg 1998; 66: 33–7.
19. Soeter JR, Smith GT, Anema RJ, et al. Distribution of oxygenated blood in femoral and brachial artery perfusion during venoarterial bypass in primates. J Thorac Cardiovasc Surg 1973; 65: 825–9.
20. The acute respiratory distress syndrome network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342: 1301–8.
21. Kinsella JP, Gerstmann DR, Rosenberg AA. The effect of extracorporeal membrane oxygenation on coronary perfusion and regional blood flow distribution. Pediatr Res 1992; 31: 80–4.
22. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000; 342: 1334–49.
23. Muscedere JG, Mullen JBM, Gan K, Slutsky AS. Tidal ventilation at low airway pressures can augment lung injury. Am J Respir Crit Care Med 1994; 149: 1327–34.
24. Ranieri VM, Suter PM, Tortorella C, et al. Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome: a randomized controlled trial. JAMA 1999; 282: 54–61.
25. Amato MBP, Barbas CSV, Medeiros DM, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338: 347–54.
© 2001 International Anesthesia Research Society