Massive pulmonary hemorrhage is a potentially fatal complication of pulmonary thromboendarterectomy (PTE). Although briefly mentioned in the medical literature (1,2), pulmonary bleeding as a surgical complication of PTE has not been discussed in detail. We report three cases in which uncontrolled bleeding from the lungs occurred immediately after cardiopulmonary bypass (CPB) and PTE. The management of post-CPB pulmonary bleeding and contributing factors unique to PTE are presented.
A 78-yr-old woman with chronic thromboembolic pulmonary hypertension (CTEPH) presented for PTE. The diagnosis of CTEPH was based on exercise intolerance and on echocardiographic and pulmonary angiographic findings. Her medical history was remarkable only for chronic systemic hypertension. Preoperative medications included metoprolol, digoxin, and furosemide. The anesthetic induction and pre-CPB period were uneventful. While performing the endarterectomy, which was performed with two episodes of deep hypothermic circulatory arrest (DHCA) (one for each pulmonary artery (PA)), the surgeon noted that the vessels were friable. During rewarming, the perfusionist encountered difficulty maintaining intravascular volume and, in addition to providing multiple boluses of crystalloid and colloid solution, administered 3 U of homologous packed red blood cells. Upon attempted separation from CPB, when mechanical ventilation and pulmonary blood flow were established, large amounts of dark blood appeared in the endotracheal tube and anesthesia circuit. The tube was suctioned, and 2 L of blood was removed. Because the bleeding was profuse and continuous, CPB was reinstituted, and the PA vent catheter was reactivated. Under direct laryngoscopy, the endotracheal tube was replaced with a size 37 left-sided double-lumen endobronchial tube. Fiberoptic bronchoscopy was performed while pulmonary blood flow was gradually allowed to resume in an attempt to identify the source of bleeding. Diffuse bleeding was noted from both sides of the tracheobronchial tree. A second attempt at separation from CPB was then initiated. This involved 10 cm H2O positive end-expiratory pressure (PEEP), the administration of 20 U arginine vasopressin diluted to 10 mL with normal saline into the tracheobronchial tree via the endobronchial tube, protamine, 4 U of fresh frozen plasma, and 10 U of platelets. Also, intermittent attempts at one-lung ventilation were applied in an attempt to determine whether bleeding was more severe on one side versus the other. These maneuvers were unsuccessful, and CPB was reinstituted after reheparinization. Repair of the PAs was deemed impossible, and, after discussions with the patients’ family, heroic efforts were withdrawn, and she was allowed to die. Postmortem examination revealed diffuse bilateral hemorrhage with multiple areas of vascular disruption.
A 76-yr-old man presented for PTE and coronary artery bypass grafting. His CTEPH was secondary to factor V Leiden, and his coronary artery disease consisted of a 70% occlusion of the left anterior descending artery (LAD). Anesthetic induction and the pre-CPB period were uneventful, as was the pulmonary endarterectomy. The endarterectomy was performed with two episodes of DHCA. The coronary bypass (left internal mammary artery to LAD) was performed early in the rewarming phase. As in Case 1, upon separation from CPB, large amounts of blood appeared in the anesthesia circuit; at least 2 L of blood was suctioned. Because of continued bleeding and difficulty with ventilation and oxygenation, CPB was reinstituted. A size 41 left-sided double-lumen endobronchial tube was placed over a tube changer stylet. Its proper position was confirmed by fiberoptic bronchoscopy, which showed bleeding from both sides, but mostly from the left lower lobe bronchus. The surgeon was able to identify and oversew a source of bleeding in the right PA but was unable to access the left lower lobe artery for repair. Consequently, 10 cm H2O of PEEP was applied to both lungs, and the left lower lobe bronchus was occluded with a Fogarty balloon catheter. The patient was transferred to the intensive care unit (ICU), where he initially improved considerably. Lung isolation no longer being necessary, the Fogarty catheter was removed, and a single-lumen endotracheal tube was inserted on the second postoperative day. Unfortunately, the patient experienced multiple organ system failure and residual pulmonary hypertension refractory to pulmonary vasodilator therapy. He died 1 wk after the operation.
A 58-yr-old woman presented for PTE. The preoperative pulmonary angiogram was consistent with the diagnosis of CTEPH, but it also indicated the likely presence of significant small-vessel (arteriolar) disease. The patient was informed of the risk of residual pulmonary hypertension posed by the arteriolar narrowing. However, because of severe exercise intolerance, she elected to receive surgery. The operative course was uneventful, but massive pulmonary bleeding (800 mL) began upon an unsuccessful attempt at separation from CPB. During a second attempt at separation from CPB, PEEP 10 cm H2O was applied; phenylephrine 10 mg and vasopressin 20 U diluted to 10 mL with normal saline were administered via the endotracheal tube; and protamine, 2 U of fresh frozen plasma, and 10 U of platelets were given. The bleeding subsided, but soon copious amounts of pink frothy sputum suggestive of pulmonary edema appeared in the endotracheal tube. The patient was brought to the ICU requiring significant inotropic support (dopamine 10 μg kg−1 min−1 and epinephrine 0.1 μg kg−1 min−1) and was ventilator dependent with reperfusion lung injury for 15 days. She was discharged after a prolonged hospital stay, with substantial functional and hemodynamic improvement.
Massive pulmonary hemorrhage after PTE is a rare event in our institution. The 3 cases presented constitute the total number of such instances in our most recent cohort of 600 PTEs (incidence 3/600 = 0.5%). This is a significant decrease in incidence compared with that reported for an earlier cohort by a different surgical team (2%) (1). This is remarkable considering that the operation is technically demanding and involves a complete endarterectomy of the entire pulmonary vascular tree. Through a median sternotomy, endarterectomies are performed on each PA by using DHCA. Typically, DHCA is performed twice—once for each side of the pulmonary vascular tree—interrupted by a 10-minute period of perfusion. DHCA provides for a bloodless surgical field, interrupting both PA and bronchial artery flow. For each PA, a 1- to 2-cm incision is made proximally, an endarterectomy plane is raised, a long blunt suction/cuvette is used to separate the fibrinous material from the vessel wall, and gentle traction on the fibrinous material is simultaneously applied with forceps. The plane is gradually extended into segmental and subsegmental vessels, with limited access and visibility. The most important factor determining the incidence of this complication is probably the experience and skill of the surgeons. The two surgeons performing PTE at University of California, San Diego (SWJ and DPK), have collectively performed more than 1400 PTEs.
Patient risk factors for this complication are difficult to identify, although the friability of the vessels in Case 1 and the advanced age in Cases 1 and 2 probably contributed. Also, the presence of residual pulmonary hypertension likely exacerbates bleeding.
The management of pulmonary hemorrhage after PTE involves treatment modalities applied to pulmonary bleeding from other causes, such as PA rupture from the PA catheter, spontaneous rupture, and trauma. Bleeding from intraoperative disruption of PAs (surgical or PA catheter) usually manifests when PA blood flow is resumed at the conclusion of CPB, the heart is allowed to eject, and ventilation resumes. Unfortunately, during PTE, testing the integrity of the PAs after each endarterectomy is not feasible. Because at that time the patient is profoundly hypothermic and the heart and great vessels are empty, cardiac ejection of blood into the PA with production of a representative systolic blood pressure would probably be necessary to perform such a test.
A general algorithm for treatment of post-CPB pulmonary hemorrhage is presented in Figure 1. The two main goals are to prevent exsanguination and ensure adequate gas exchange. If adequate hemodynamic status and gas exchange can be maintained, conservative management consisting of positive airway pressure, topical vasoconstrictors, reversal of heparin, and correction of coagulopathies can be provided. If not, resumption of CPB is indicated, and a diagnostic bronchoscopy should be performed to identify the source of bleeding. Lung isolation may be useful in unilateral bleeding, with lobar endobronchial blockade as in Case 2 (3,4). PEEP may tamponade the bleeding (5). Other approaches to pulmonary hemorrhage have involved vascular occlusion with balloon inflation (6), banding (7), or embolization (8–10) and pulmonary resection (11). These were not viable options in Cases 1 and 2 because of the diffuse nature of the bleeding, and they were not necessary in Case 3. In two of the cases, topical vasoconstrictors (vasopressin and phenylephrine) were administered. In one case (Case 3), this maneuver may have contributed to success. Topical administration of vasoconstrictors such as vasopressin analogs and epinephrine decreases endobronchial bleeding in the ICU setting (12,13), as well as during diagnostic bronchoscopy (14). Although we were concerned about the possibility of systemic absorption of the vasoconstrictors leading to systemic and pulmonary hypertension, we did not see such responses in these cases. We chose “pure” vasoconstrictors as opposed to epinephrine to avoid the tachycardia that has been associated with tracheobronchial epinephrine administration (15).
Two factors unique to PTE, if present, complicate diagnosis and management: residual pulmonary hypertension and reperfusion pulmonary edema. The differential diagnosis of red fluid emanating from the endotracheal tube after PTE consists mainly of two conditions: bleeding and pulmonary edema. Visual inspection usually allows one to distinguish between them, but occasionally, as in Case 3, both may exist simultaneously (16). The treatment for severe acute pulmonary edema after PTE is primarily supportive (17), and, in the application of PEEP, it largely overlaps that of bleeding. With pulmonary edema, however, the situation is often less desperate, and extra-corporeal membrane oxygenation can be considered as a temporizing measure during pulmonary healing.
Pulmonary hypertension should be controlled if possible. Causes of high pulmonary vascular resistance, such as hypercarbia, hypoxemia, and light anesthesia, should be treated. For refractory pulmonary hypertension, the administration of pulmonary vasodilators such as milrinone, inhaled nitric oxide, and inhaled prostacyclin may be considered. As seen in Case 2, however, the available pulmonary vasodilators may not be effective in controlling residual pulmonary hypertension after PTE (18).
PTE is currently being performed with increasing frequency at our institution and throughout the world (19). It is now widely recognized as the preferred treatment of CTEPH, having supplanted pulmonary transplantation because of its decreased rate of peri-operative mortality and better long-term outcome (17,19). Pulmonary hemorrhage, although rare, is the third most common cause for perioperative mortality in our PTE population, with residual pulmonary hypertension and reperfusion pulmonary edema the leading causes (19). Anesthesiologists involved in PA surgery must be fully prepared to provide diagnostic and therapeutic maneuvers such a bronchial blockade, differential lung positive pressure, endobronchial administration of vasoconstrictors, and pharmacologic control of residual pulmonary hypertension. Other options, such as embolization and pulmonary resection, may also be considered in selected circumstances.
1. Daily PO, Dembitzsky WP, Iversen S, et al. Current early results of pulmonary thromboendarterectomy for chronic pulmonary embolism. Eur J Cardiothorac Surg 1990;4:117–21; discussion 122–3.
2. Chandler J, Grixti K, Dunning J, Vuylsteke A. An unusual chest radiograph after pulmonary thromboendarterectomy. J Cardiothorac Vasc Anesth 2002;16:253–4.
3. Mangar D, Connell GR, Lessin JL, Rasanen J. Catheter-induced pulmonary artery haemorrhage resulting from a pneumothorax. Can J Anaesth 1993;40:1069–72.
4. Purut CM, Scott SM, Parham JV, Smith PK. Intraoperative management of severe endobronchial hemorrhage. Ann Thorac Surg 1991;51:304–6; discussion 306–7.
5. Sumita S, Ujike Y, Namiki A, et al. Rupture of pulmonary artery induced by balloon occlusion pulmonary angiography. Intensive Care Med 1995;21:79–81.
6. Bredin CP, Richardson PR, King TK, et al. Treatment of massive hemoptysis by combined occlusion of pulmonary and bronchial arteries. Am Rev Respir Dis 1978;117:969–73.
7. Stone JG, Faltas AN, Khambatta HJ, et al. Temporary unilateral pulmonary artery occlusion: a method for controlling Swan-Ganz catheter-induced hemoptysis. Ann Thorac Surg 1984;37: 508–10.
8. Carlson TA, Goldenberg IF, Murray PD, et al. Catheter-induced delayed recurrent pulmonary artery hemorrhage: intervention with therapeutic embolism of the pulmonary artery. JAMA 1989;261:1943–5.
9. Jondeau G, Lacombe P, Rocha P, et al. Swan-Ganz catheter-induced rupture of the pulmonary artery: successful early management by transcatheter embolization. Cathet Cardiovasc Diagn 1990;19:202–4.
10. Tayoro J, Dequin PF, Delhommais A, et al. Rupture of pulmonary artery induced by Swan-Ganz catheter: success of coil embolization. Intensive Care Med 1997;23:198–200.
11. Sekkal S, Cornu E, Christides C, et al. Swan-Ganz catheter induced pulmonary artery perforation during cardiac surgery concerning two cases. J Cardiovasc Surg (Torino) 1996;37:313–7.
12. Sidman JD, Wheeler WB, Cabalka AK, et al. Management of acute pulmonary hemorrhage in children. Laryngoscope 2001; 111:33–5.
13. Dupree HJ, Lewejohann JC, Gleiss J, et al. Fiberoptic bronchos-copy of intubated patients with life-threatening hemoptysis. World J Surg 2001;25:104–7.
14. Breuer HW, Charchut S, Worth H, et al. Endobronchial versus intravenous application of the vasopressin derivative glypressin during diagnostic bronchoscopy. Eur Respir J 1989;2:225–8.
15. Efrati O, Barak A, Ben-Abraham R, et al. Should vasopressin replace adrenaline for endotracheal drug administration? Crit Care Med 2003;31:572–6.
16. Lee KC, Cho YL, Lee SY. Reperfusion pulmonary edema after pulmonary endarterectomy. Acta Anaesthesiol Sin 2001;39: 97–101.
17. Fedullo PF, Auger WR, Kerr KM, Rubin LJ. Chronic thromboembolic pulmonary hypertension. N Engl J Med 2001;345: 1465–72.
18. Fedullo PF, Auger WR, Dembitsky WP. Postoperative management of the patient undergoing pulmonary thromboendarterectomy. Semin Thorac Cardiovasc Surg 1999;11:172–8.
19. Jamieson SW, Kapelanski DP, Sakakibara N, et al. Pulmonary endarterectomy: experience and lessons learned in 1,500 cases. Ann Thorac Surg 2003;76:1457–64.