- Email: [email protected]
Balloon catheter tamponade in cardiovascular wounds
Balloon Catheter Tamponade in Cardiovascular Wounds David V. Feliciano, MD, Jon M. Butch, MD, Kenneth L. Mattox, MD, Carmel G. Bitondo, MD, Gwen Fields, MD, Houston,Texas
From 1980 to 1990, operative balloon catheter tamponade was used in 12 patients with cardiac or vascular injuries from penetrating wounds. In nine patients, a balloon catheter was passed into a bleeding site through a bullet track or proximal artery and inflated with saline or radiologic dye. In two of these patients, the proximal balloon catheter was folded on itself, tied in that position, and left in the patient permanently. In the other seven patients in this group, the balloon catheter was attached to a three-way stopcock and left temporarily inflated postoperatively. These patients were then observed in the intensive care unit for 3 to 4 days, at which time the balloon was withdrawn. Eight of nine patients survived without recurrent hemorrhage after removal of the balloon catheter, while one patient with a Fogarty balloon placed in the carotid siphon died of a cerebral infarction. Balloon catheter tamponade was also used on a temporary basis in one patient with a posterior cardiac w o u n d and in one patient with an anterior stab wound of the inferior vena cava at the renal veins, whereas in two patients with high cervical arteriovenous fistulas, one had permanent placement of the balloon catheter while the other had temporary placement. One of the latter patients also had acute hemorrhage. Although all four patients survived, one of the patients with a fistula developed a recurrence and another required two separate operative procedures for correct placement of the balloon to cure the fistula.
From the Cora and Webb Mading Department of Surgery, Baylor College of Medicine, and the Ben Taub General Hospital, Houston, Texas. Requests for reprints should be addressed to David V. Feliciano, MD, Department of Surgery, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, New York 14642. Presented at the 42nd Annual Meeting of the Southwestern Surgical Congress, La Quinta, California, April 22-25, 1990.
Perative balloon catheter tamponade is a useful technique to control intraoperative hemorrhage from inaccessible or fragile cardiovascular structures. First used by Taylor and Williams [1] in 1960, Foley or Fogarty balloon catheters have since been utilized in a variety of elective and emergent cardiac and vascular operations [2-7]. Herein, we describe a 10-year experience with balloon catheter tamponade in traumatic cardiovascular wounds. All patients were treated at the Ben Taub General Hospital, a level I trauma center affiliated with the Baylor College of Medicine in Houston, Texas.
O
PATIENTS AND M E T H O D S The records of 12 patients who had a balloon catheter used for tamponade in a cardiac or vascular wound between January 1980 and January 1990 were reviewed. Evaluation and management of patients with cardiac or vascular wounds were reasonably consistent during this period [8-10]. Standard techniques of cardiorrhaphy were used during the 10-year interval of the review [9]. On rare occasions, more advanced techniques of repair utilizing adjuncts such as balloon catheter tamponade, inflow occlusion, or cardiopulmonary bypass were necessary. Balloon catheter tamponade was required only once during this period in a patient with a cardiac wound who did not have a prehospital cardiopulmonary arrest and therefore had a reasonable chance for survival. In patients with obvious external exsanguinating hemorrhage from a vascular injury caused by a penetrating wound, control was attempted with a finger or compression dressing in the emergency center prior to transfer to the operating room. The decision to insert a balloon catheter in the operating room was based on the persistence of exsanguinating hemorrhage and the inaccessible location of or difficulty in controlling the injured vessel. A no. 3 to no. 8 Fogarty balloon catheter or 30-mL Foley balloon catheter was inserted directly into the entrance wound of the missile or knife track, directly into the defect in the injured vessel, or into the proximal part of an injured artery, if it could be exposed quickly. The balloon was then advanced and periodically inflated until the appropriate location was reached and hemorrhage ceased. In patients in whom the injured vessel could then be directly visualized, repair was performed and the balloon catheter was withdrawn in the operating room. In most of the remaining patients, inflation of the balloon was performed with saline or radiologic dye, depending on the location of the balloon and the desire for visualization of the balloon on postoperative roentgenograms. A threeway stopcock was also attached to the catheter to prevent
THE AMERICAN J O U R N A L O F SURGERY
VOLUME 160
DECEMBER 1990
583
FELIC|ANOET AL
Figure 1. A no. 8 Fogarty balloon catheter was inserted in high cervical stab wound to control exsanguinating hemorrhage. Deflation of the balloon and removal of the catheter were performed on the fourth day after insertion.
deflation of the balloon in the postoperative period. In several patients, gauze packing was used to fill the track of the missile or knife and hold the balloon catheter in place. Deflation of the balloon and removal of the catheter were performed 3 to 4 days after insertion, occasionally in the operating room, but most commonly at the bedside in the surgical intensive care unit. In two patients, the proximal balloon catheter was folded upon itself, ties were placed around this area to keep the catheter folded, the female attachment at the end of the catheter was cut off, and the folded end was placed in the soft tissue of the neck prior to closure. Balloon catheters were also used in two instances to occlude high cervical arteriovenous fistulas secondary to gunshot wounds. The technique involved passage of the catheter up the injured artery and periodic inflation until the palpable thrill or audible bruit associated with the arteriovenous fistula disappeared. The proximal end of the balloon catheter was then folded upon itself and left in place as described previously. All patients in the series were treated by one of the three attending surgeons (DVF, JMB, KLM) listed as authors. Data were also obtained from individual patient charts and included demographics of the patient population, indication for use of a balloon catheter, type of balloon catheter used, location of the balloon catheter, day of removal, and postoperative results. RESULTS From January 1980 to January 1990, 12 patients (11 men and 1 woman, average age: 26 years) had a balloon catheter inserted for tamponade of a cardiac or vascular wound. Penetrating wounds were the mechanism of injury in all 12 patients and included 10 gunshot wounds and 2 stab wounds. The indication for use of a balloon cathe584
ter was exsanguinating hemorrhage from the heart in 1 patient; exsanguinating hemorrhage from an "inaccessible" artery or "difficult to control" vein in the head, neck, or trunk in 10 patients; and a high cervical arteriovenous fistula in 2 patients, with 1 patient having separate balloon catheters placed for both of the last-named indications. In the patient with a cardiac injury from a gunshot wound, tamponade with the 5-mL balloon of a Foley catheter was used in a penetrating wound to the right superior pulmonary vein at its junction with the left atrium until atriorrhaphy and venorrhaphy could be completed in a bloodless field. Despite associated wounds to the superior vena cava and left ventricle, the patient survived. In the 10 patients in whom the balloon catheter was used to control exsanguinating arterial (9 patients) or venous (1 patient) hemorrhage, the head or neck was the location in 8 patients (lumen of carotid artery at base of skull or in pharynx in 3, lumen of vertebral artery at base of skull in 1, base of skull in 1, pharynx in 1, face in 1, upper neck in 1), while 2 patients required operative placement of a balloon into the juxtarenal inferior vena cava and a bleeding pelvic artery, respectively. No. 3 or no. 4 Fogarty balloon catheters were the most commonly used size and type in this group. Two patients underwent permanent insertion of the balloon catheter, one patient had intraoperative insertion and removal of a 30-mL Foley balloon catheter to control an anterior wound in the juxtarenal inferior vena cava, and seven patients had the balloon catheter removed, most commonly on the fourth day after insertion. Nine patients in the group of 10 survived, including seven who had no recurrence of hemorrhage after delayed deflation of the balloon (Figure 1). The only death in the series occurred in a 40-year-old man in this group who had insertion of a no. 4 Fogarty balloon catheter into the carotid siphon to control exsanguinating hemorrhage from the base of the skull. He was comatose postoperatively but moved all four extremities. His intracranial pressure was monitored and gradually increased, while a computed tomographic scan revealed evidence of a right middle cerebral infarct with diffuse swelling of the brain. The patient died on the sixth postoperative day with the balloon catheter in place. Two patients underwent insertion of no. 3 Fogarty balloon catheters to control high cervical arteriovenous fistulas, including one who had survived the insertion of two 30-mL Foley balloon catheters into the nasopharynx to control exsanguinating hemorrhage caused by a gunshot wound. This patient was noted to have a high carotid-jugular arteriovenous fistula 7 days after the second 30-mL Foley balloon catheter was removed from the nasopharynx. At a subsequent cervical operation, inflation of the no. 3 Foley balloon catheter in the right internal carotid artery below the fistula caused significant electrocardiographic changes; therefore, a right superficial temporal artery-middle cerebral artery bypass was performed before the balloon in the distal carotid artery was reinflated at the site of the fistula. The catheter was then folded over and left in the neck. An ipsilateral carot-
THE AMERICAN JOURNAL OF SURGERY VOLUME160 DECEMBER1990
BALLOONCATHETERTAMPONADEIN CARDIOVASCULARWOUNDS
id arteriogram revealed no flow through the fistula, while a contralateral carotid arteriogram 9 days later showed persistence of the fistula above the previously placed no. 3 Fogarty balloon. A reoperation with passage of a new no. 3 Fogarty balloon into the carotid siphon above the recurrent fistula led to a cure (Figure 2). Operative placement of a no. 3 Fogarty balloon catheter into a high vertebraljugular arteriovenous fistula in another patient was curative for 24 days, but the fistula recurred with removal of the balloon. A direct operative approach was subsequently curative. In summary, both of the first attempts at operative placement of a balloon catheter to control a high cervical arteriovenous fistula failed in this series. COMMENTS Routine methods for control of hemorrhage are applicable in most patients with penetrating cardiac or vascular wounds. There is, however, a small subset of patients with exsanguinating wounds that are too large, fragile, or inaccessible to allow for the use of vascular clamps, vascular tapes or loops, or rapid suture techniques. In this group of patients, either temporary or permanent placement of a balloon catheter will allow for cardiac or vascular repair in a relatively bloodless field. After the previously mentioned first report of intraoperative balloon catheter tamponade of an iliac arteriovenous fistula, subsequent reports documented the efficacy of the technique in penetrating wounds to the head and neck [7]; mishaps during elective operations in the head and neck [6,11]; penetrating cardiac wounds [2]; mishaps during elective cardiac operations [12]; penetrating abdominal vascular wounds [13]; emergency abdominal vascular procedures, such as that for ruptured abdominal aortic aneurysms [14]; and penetrating wounds of the deep pelvis [5,15]. Prior to the availability of interventional radiologists and percutaneous detachable balloons [16,17], carotid-cavernous, carotid-jugular, and vertebral-jugular arteriovenous fistulas were also treated with operative placement of intra-arterial balloon catheters
[4,18-22]. In patients with perforations or rupture of the heart, vascular clamps can be used on the atria and finger compression or tamponade on the ventricles, as patients still alive on admission usually have limited injuries. With larger injuries in the ventricles, the previously mentioned inflow occlusion (clamping of the superior and inferior vena cavae) or pharmacologic hypotension may allow for the rapid placement of several deep mattress sutures to close the defect [23]. An alternative approach is to insert a Foley catheter with either a 5-mL or 30-mL balloon into the defect, inflate the balloon, and put traction on the catheter so that the balloon fills the defect [12,23,24]. A cuffed endotracheal tube has been used for the same purpose in one report [25]. A pursestring suture can then be placed around the margins of the defect and pulled up tight as the balloon is deflated and the catheter removed. Besides large cardiac injuries, balloon catheter tamponade is also useful when fragile or inaccessible areas of the heart and central circulation are injured. Examples in the former group would include a thinned-out wall of the
Figure 2. A no. 3 Fogarty balloon catheter containing radiologic dye is visualized in the intracranial carotid artery. The balloon catheter was left in place to prevent recurrence of a high carotidjugular arteriovenous fistula,
right ventricle, the pulmonary artery, and the pulmonary veins. In the latter group, the posterior wall of the right atrium at the caval junctions, posterior wall of the left atrium at the junction with the pulmonary veins, the high posterior wall of the ventricles, and the posterior wall of the ascending aorta and transverse arch are all difficult to occlude with a finger and suture when the heart is still contracting. In each instance, insertion of a balloon catheter may be the quickest and safest technique to control exsanguinating hemorrhage as a decision is reached on the most appropriate technique of cardiorrhaphy or aortorrhaphy. The primary indication for use of balloon catheter tamponade in noncardiac wounds has been the inaccessible location of the injured artery. The carotid artery behind the pharynx or at the base of the skull; the internal maxillary artery and its branches deep in the face; the second, third, and fourth positions of the vertebral artery [26,27]; the internal jugular vein at the base of the skull [11]; and the distal branches of the internal iliac artery in the pelvis are all vessels embedded in or protected by bony structures. Passing a balloon catheter, usually a Fogartytype [28], into the track of a missile or knife wound or into the proximal artery (or distal vein) and then advancing the catheter with periodic inflation will usually allow for control of hemorrhage at the site of injury, despite the lack of classic proximal and distal vascular control. The distal end of the catheter is then attached to a three-way stopcock, the syringe removed, and the catheter left hanging out of the track or one end of the incision. Originally,
THE AMERICAN JOURNAL OF SURGERY VOLUME160 DECEMBER1990 585
FELICIANOET AL
patients in this review were returned to the operating room 3 to 4 days after insertion of the balloon catheter so that deflation could be performed under ideal circumstances. When hemorrhage was noted to be controlled in all instances in the first several patients, subsequent deflations were performed in the surgical intensive care unit and the catheter was removed without a return to the operating room. A n alternative approach is to fold the end of t h e proximal catheter, tie it in this position, and leave it in place, as was described for the patients with cervical arteriovenous fistulas [4]. A difficult problem is the development of ipsilateral cerebral ischemia after balloon occlusion of the high cervical internal carotid artery. If this diagnosis is obvious based on an intraoperative or postoperative electroencephalogram or postoperative computed tomographic scan, rapid performance of an extracranial-intracranial bypass may prevent or reverse neurologic deficits [29]. The one death in this series might have been prevented by such an aggressive approach to cerebral revascularization. The dismal results with operative placement of balloon catheters to control high cervical arteriovenous fistulas in this review suggest that radiologic placement under fluoroscopic control is a better approach. In truth, the patients reported underwent operations in 1980 and 1982, and radiologic techniques have been utilized in our institution since that time, much as in other centers [30]. If operative placement of a balloon catheter to control a high cervical arteriovenous fistula becomes necessary, admittedly an unlikely event in 1990, the technique is as follows: the proximal artery is ligated, the balloon catheter is passed into the distal artery beyond the tie, and the balloon is inflated when the measured level of the fistula is reached or the bruit is noted to disappear after inflation. Inflation of the balloon with radiologic dye will allow for easier visualization on subsequent arteriograms, as previously noted. A n intraoperative cerebral arteriogram is then performed through the contralateral carotid or vertebral artery to verify that retrograde flow through intracerebral collaterals is not keeping the fistula open above the balloon. Intraoperative electroencephalographic monitoring will help determine the need for an ipsilateral extracranial-intracranial bypass if a carotid arteryjugular vein fistula is being occluded. An appropriately placed balloon can be left in place permanently by fixing the proximal end of the catheter in a folded position as described [4]. Balloon catheter tamponade, which has now been performed for 30 years, remains a useful technique in selected patients with hemorrhage from cardiac or vascular injuries that are difficult to control with standard techniques. When exsanguinating hemorrhage from a large defect or a fragile or inaccessible cardiac or vascular structure is encountered, balloon catheter tamponade should be used.
REFERENCES 1. Taylor H, Williams E. Arteriovenous fistula following disk surgery. Br J Surg 1962; 50: 47-50.
586
2. Pearce CW, McCool E, Schmidt FE. Control of bleeding from cardiovascular wounds: balloon catheter tamponade. Ann Surg 1966; 166: 257-9. 3. Foster JH, Morgan CV, Threlkel JB. Proximal control of aorta with a balloon catheter. Surg Gynecol Obstet 1971; 132: 693-4. 4. Binkley FM, Wylie EJ. A new technique for obliteration of cerebrovascular arteriovenous fistulae. Arch Surg 1973; 106: 5247. 5. Sheldon GF, Winestock DP. Hemorrhage from open pelvic fracture controlled intraoperatively with balloon catheter. J Trauma 1978; 18: 68-70. 6. Reilly JJ Jr, Caparosa R J, Latchaw RE, Sheptak PE. Aberrant carotid artery injured at myringotomy. Control of hemorrhage by a balloon catheter. JAMA 1983; 249: 1473-4. 7. Belldn M, Dunton R, Crombie HD Jr, Lowe R. Preoperative percutaneous intraluminal balloon catheter control of major arterial hemorrhage. J Trauma 1988; 28: 548-50. 8. Feliciano DV, Bitondo CG, Mattox KL, et al. Civilian trauma in the 1980s. A 1-year experience with 456 vascular and cardiac injuries. Ann Surg 1984; 199: 717-24. 9. Feliciano DV, Mattox KL. Indications, technique, and pitfalls of emergency center thoracotomy. Surg Rounds 1981; 4: 32-40. IO. Feliciano DV. Managing peripheral vascular trauma. Infect Surg 1986; 5: 659-69, 682. 11. Cohen IK, Chretien PB. Control of hemorrhage from the internal jugular vein by a balloon catheter. Surg Gynecol Obstet 1973; 136: 791-2. 12. Dube P. Control of massive cardiac haemorrhage by balloon catheter. Thorax 1973; 28: 399-400. 13. Smiley K, Perry MO. Balloon catheter tamponade of major vascular wounds. Am J Surg 1971; 121: 326-7. 14. Hyde GL, Sullivan DM. Fogarty catheter tamponade of ruptured abdominal aortic aneurysms. Surg Gynecol Obstet 1982; 154: 197-9. 15. Owen DR, Hodgson PE. Control of hemorrhage following missile wound to the pelvis. J Trauma 1980; 20: 906-8. 16. Barrow DL, Fleischer AS, Hoffman JC. Complications of detachable balloon catheter technique in the treatment of traumatic intracranial arteriovenous fistulas. J Neurosurg 1982; 56: 396-403. 17. Miller RE, Hieshima GB, Giannotta SL, Grinnell VS, Mehringer CM, Kerin DS. Acute traumatic vertebral arteriovenous fistula: balloon occlusion with the use of a contralateral approach. Neurosurgery 1984; 14: 225-9. 18. Proto DJ, Hanbery JW. Intraluminal occlusion of a carotidcavernous sinus fistula with a balloon catheter. Technical note. J Neurosurg 1971; 35: 237-42. 19. Scott TE Jr, Tweed CG, Hollis BF, Eardley K. Management of traumatic carotid cavernous fistula using the Fogarty catheter technique. J Trauma 1980; 20: 610-13. 20. Debrun G, Lacour P, Vinuela F, Fox A, Drake CG, Caron JP. Treatment of 54 traumatic carotid-cavernous fistulas. J Neurosurg 1981; 55: 678-91. 21. Goodman SJ, Hasso A, Kirkpatrick D. Treatment of vertebrojugular fistula by balloon occlusion. Case report. J Neurosurg 1975; 43: 362-7. 22. Fairman RM, Grossman RI, Goldberg HI, Kivuls J, Perloff LJ. A new approach to the treatment of vertebral arteriovenous fistulas. Surgery 1984; 95: 112-5. 23. Trinkle JK, Toon RS, Franz JL, Arom KV, Grover FL. Affairs of the wounded heart: penetrating cardiac wounds. J Trauma 1979; 19: 467-72. 24. Arbuln A, Thorns NW. Control of bleeding from a gunshot wound of the inferior vena cava at its junction with the right atrium by means of a Foley catheter. J Thorac Cardiovasc Surg 1972; 63: 427-9: 25. Oelert H, Dalichau H, Borst HG. The use of a cuffed endotracheal tube for repair of operative injuries to the atria. Ann Th0rac Surg 1974; 18: 102-3. 26. Meier DE, Brink BE, Fry WJ. Vertebral artery trauma. Acute
THE AMERICAN JOURNAL OF SURGERY VOLUME 160 DECEMBER 1990
BALLOONCATHETERTAMPONADEIN CARDIOVASCULARWOUNDS
recognition and treatment. Arch Surg 1981; 116: 236-9. 27. Golueke P, Sclafani S, Phillips T, Goldstein A, Scalea T, Duncan A. Vertebral artery injury--diagnosis and management. J Trauma 1987; 27: 856-65. 28. Fogarty T J, Cranley J J, Krause R J, Strassner ES, Hafner CD. A method for extraction of arterial emboli and thrombi. Surg Gynecol Obstet 1963; 116: 241-4. 29. Gewertz BL, Samson DS, Ditmore QM, Bone GE. Management of penetrating injuries of the internal carotid artery at the base of the skull utilizing extraeranial-intracranial bypass. J Trauma 1980; 20: 365-9. 30. Sclafani SJ, Panetta T, Goldstein AS, et al. The management of arterial injuries caused by penetration of zone III of the neck. J Trauma 1985; 25: 871-81.
DISCUSSION Gilbert S. Campbell (Little Rock, AR): With the tremendous experience and contributions that Dr. Feliciano and his associates have made through the years in trauma, to have utilized this particular procedure only 10 times in 10 years underscores its infrequent need. The balloon-tipped catheter tamponade technique is a lifesaving measure rarely used. Dr. Feliciano, you stated in the manuscript that the balloon-tipped catheter was left permanently in one patient. How long a period of time did you mean by "permanently?"
Richard E. Ward (Sacramento, CA): Were heparin or anticoagulants used in these patients, especially those with carotid balloon tamponade? David V. Felieiano (closing): Dr. Campbell, if the injured artery is in an inaccessible location, these catheters can be folded over and tied in that position. The balloon will stay inflated in the vessel. In most of the patients in this series, the balloons were simply deflated 3 to 4 days later in the operating room or in the surgical intensive care unit because we became convinced that bleeding would not recur with deflation. In two patients who required balloon insertion to control hemorrhage and in one patient with a balloon inserted to close a high carotid-jugular fistula, the balloons have never been removed. Dr. Ward, we did not give heparin to any patients in this series. I am not sure that this would prevent the complication of ipsilateral cerebral infarction. In a small subset of patients, the circle of Willis is not going to protect these patients if one carotid artery is occluded. The lesson in this series is that surgeons should keep this technique in mind. There are few places in the body where one of these small balloon catheters cannot be placed. If the balloon is left inflated for 3 to 4 days, the arterial tree thromboses around it, and there will be no further hemorrhage.
THE AMERICAN JOURNAL OF SURGERY VOLUME160 DECEMBER 1990 587
From 1980 to 1990, operative balloon catheter tamponade was used in 12 patients with cardiac or vascular injuries from penetrating wounds. In nine patients, a balloon catheter was passed into a bleeding site through a bullet track or proximal artery and inflated with saline or radiologic dye. In two of these patients, the proximal balloon catheter was folded on itself, tied in that position, and left in the patient permanently. In the other seven patients in this group, the balloon catheter was attached to a three-way stopcock and left temporarily inflated postoperatively. These patients were then observed in the intensive care unit for 3 to 4 days, at which time the balloon was withdrawn. Eight of nine patients survived without recurrent hemorrhage after removal of the balloon catheter, while one patient with a Fogarty balloon placed in the carotid siphon died of a cerebral infarction. Balloon catheter tamponade was also used on a temporary basis in one patient with a posterior cardiac w o u n d and in one patient with an anterior stab wound of the inferior vena cava at the renal veins, whereas in two patients with high cervical arteriovenous fistulas, one had permanent placement of the balloon catheter while the other had temporary placement. One of the latter patients also had acute hemorrhage. Although all four patients survived, one of the patients with a fistula developed a recurrence and another required two separate operative procedures for correct placement of the balloon to cure the fistula.
From the Cora and Webb Mading Department of Surgery, Baylor College of Medicine, and the Ben Taub General Hospital, Houston, Texas. Requests for reprints should be addressed to David V. Feliciano, MD, Department of Surgery, University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, New York 14642. Presented at the 42nd Annual Meeting of the Southwestern Surgical Congress, La Quinta, California, April 22-25, 1990.
Perative balloon catheter tamponade is a useful technique to control intraoperative hemorrhage from inaccessible or fragile cardiovascular structures. First used by Taylor and Williams [1] in 1960, Foley or Fogarty balloon catheters have since been utilized in a variety of elective and emergent cardiac and vascular operations [2-7]. Herein, we describe a 10-year experience with balloon catheter tamponade in traumatic cardiovascular wounds. All patients were treated at the Ben Taub General Hospital, a level I trauma center affiliated with the Baylor College of Medicine in Houston, Texas.
O
PATIENTS AND M E T H O D S The records of 12 patients who had a balloon catheter used for tamponade in a cardiac or vascular wound between January 1980 and January 1990 were reviewed. Evaluation and management of patients with cardiac or vascular wounds were reasonably consistent during this period [8-10]. Standard techniques of cardiorrhaphy were used during the 10-year interval of the review [9]. On rare occasions, more advanced techniques of repair utilizing adjuncts such as balloon catheter tamponade, inflow occlusion, or cardiopulmonary bypass were necessary. Balloon catheter tamponade was required only once during this period in a patient with a cardiac wound who did not have a prehospital cardiopulmonary arrest and therefore had a reasonable chance for survival. In patients with obvious external exsanguinating hemorrhage from a vascular injury caused by a penetrating wound, control was attempted with a finger or compression dressing in the emergency center prior to transfer to the operating room. The decision to insert a balloon catheter in the operating room was based on the persistence of exsanguinating hemorrhage and the inaccessible location of or difficulty in controlling the injured vessel. A no. 3 to no. 8 Fogarty balloon catheter or 30-mL Foley balloon catheter was inserted directly into the entrance wound of the missile or knife track, directly into the defect in the injured vessel, or into the proximal part of an injured artery, if it could be exposed quickly. The balloon was then advanced and periodically inflated until the appropriate location was reached and hemorrhage ceased. In patients in whom the injured vessel could then be directly visualized, repair was performed and the balloon catheter was withdrawn in the operating room. In most of the remaining patients, inflation of the balloon was performed with saline or radiologic dye, depending on the location of the balloon and the desire for visualization of the balloon on postoperative roentgenograms. A threeway stopcock was also attached to the catheter to prevent
THE AMERICAN J O U R N A L O F SURGERY
VOLUME 160
DECEMBER 1990
583
FELIC|ANOET AL
Figure 1. A no. 8 Fogarty balloon catheter was inserted in high cervical stab wound to control exsanguinating hemorrhage. Deflation of the balloon and removal of the catheter were performed on the fourth day after insertion.
deflation of the balloon in the postoperative period. In several patients, gauze packing was used to fill the track of the missile or knife and hold the balloon catheter in place. Deflation of the balloon and removal of the catheter were performed 3 to 4 days after insertion, occasionally in the operating room, but most commonly at the bedside in the surgical intensive care unit. In two patients, the proximal balloon catheter was folded upon itself, ties were placed around this area to keep the catheter folded, the female attachment at the end of the catheter was cut off, and the folded end was placed in the soft tissue of the neck prior to closure. Balloon catheters were also used in two instances to occlude high cervical arteriovenous fistulas secondary to gunshot wounds. The technique involved passage of the catheter up the injured artery and periodic inflation until the palpable thrill or audible bruit associated with the arteriovenous fistula disappeared. The proximal end of the balloon catheter was then folded upon itself and left in place as described previously. All patients in the series were treated by one of the three attending surgeons (DVF, JMB, KLM) listed as authors. Data were also obtained from individual patient charts and included demographics of the patient population, indication for use of a balloon catheter, type of balloon catheter used, location of the balloon catheter, day of removal, and postoperative results. RESULTS From January 1980 to January 1990, 12 patients (11 men and 1 woman, average age: 26 years) had a balloon catheter inserted for tamponade of a cardiac or vascular wound. Penetrating wounds were the mechanism of injury in all 12 patients and included 10 gunshot wounds and 2 stab wounds. The indication for use of a balloon cathe584
ter was exsanguinating hemorrhage from the heart in 1 patient; exsanguinating hemorrhage from an "inaccessible" artery or "difficult to control" vein in the head, neck, or trunk in 10 patients; and a high cervical arteriovenous fistula in 2 patients, with 1 patient having separate balloon catheters placed for both of the last-named indications. In the patient with a cardiac injury from a gunshot wound, tamponade with the 5-mL balloon of a Foley catheter was used in a penetrating wound to the right superior pulmonary vein at its junction with the left atrium until atriorrhaphy and venorrhaphy could be completed in a bloodless field. Despite associated wounds to the superior vena cava and left ventricle, the patient survived. In the 10 patients in whom the balloon catheter was used to control exsanguinating arterial (9 patients) or venous (1 patient) hemorrhage, the head or neck was the location in 8 patients (lumen of carotid artery at base of skull or in pharynx in 3, lumen of vertebral artery at base of skull in 1, base of skull in 1, pharynx in 1, face in 1, upper neck in 1), while 2 patients required operative placement of a balloon into the juxtarenal inferior vena cava and a bleeding pelvic artery, respectively. No. 3 or no. 4 Fogarty balloon catheters were the most commonly used size and type in this group. Two patients underwent permanent insertion of the balloon catheter, one patient had intraoperative insertion and removal of a 30-mL Foley balloon catheter to control an anterior wound in the juxtarenal inferior vena cava, and seven patients had the balloon catheter removed, most commonly on the fourth day after insertion. Nine patients in the group of 10 survived, including seven who had no recurrence of hemorrhage after delayed deflation of the balloon (Figure 1). The only death in the series occurred in a 40-year-old man in this group who had insertion of a no. 4 Fogarty balloon catheter into the carotid siphon to control exsanguinating hemorrhage from the base of the skull. He was comatose postoperatively but moved all four extremities. His intracranial pressure was monitored and gradually increased, while a computed tomographic scan revealed evidence of a right middle cerebral infarct with diffuse swelling of the brain. The patient died on the sixth postoperative day with the balloon catheter in place. Two patients underwent insertion of no. 3 Fogarty balloon catheters to control high cervical arteriovenous fistulas, including one who had survived the insertion of two 30-mL Foley balloon catheters into the nasopharynx to control exsanguinating hemorrhage caused by a gunshot wound. This patient was noted to have a high carotid-jugular arteriovenous fistula 7 days after the second 30-mL Foley balloon catheter was removed from the nasopharynx. At a subsequent cervical operation, inflation of the no. 3 Foley balloon catheter in the right internal carotid artery below the fistula caused significant electrocardiographic changes; therefore, a right superficial temporal artery-middle cerebral artery bypass was performed before the balloon in the distal carotid artery was reinflated at the site of the fistula. The catheter was then folded over and left in the neck. An ipsilateral carot-
THE AMERICAN JOURNAL OF SURGERY VOLUME160 DECEMBER1990
BALLOONCATHETERTAMPONADEIN CARDIOVASCULARWOUNDS
id arteriogram revealed no flow through the fistula, while a contralateral carotid arteriogram 9 days later showed persistence of the fistula above the previously placed no. 3 Fogarty balloon. A reoperation with passage of a new no. 3 Fogarty balloon into the carotid siphon above the recurrent fistula led to a cure (Figure 2). Operative placement of a no. 3 Fogarty balloon catheter into a high vertebraljugular arteriovenous fistula in another patient was curative for 24 days, but the fistula recurred with removal of the balloon. A direct operative approach was subsequently curative. In summary, both of the first attempts at operative placement of a balloon catheter to control a high cervical arteriovenous fistula failed in this series. COMMENTS Routine methods for control of hemorrhage are applicable in most patients with penetrating cardiac or vascular wounds. There is, however, a small subset of patients with exsanguinating wounds that are too large, fragile, or inaccessible to allow for the use of vascular clamps, vascular tapes or loops, or rapid suture techniques. In this group of patients, either temporary or permanent placement of a balloon catheter will allow for cardiac or vascular repair in a relatively bloodless field. After the previously mentioned first report of intraoperative balloon catheter tamponade of an iliac arteriovenous fistula, subsequent reports documented the efficacy of the technique in penetrating wounds to the head and neck [7]; mishaps during elective operations in the head and neck [6,11]; penetrating cardiac wounds [2]; mishaps during elective cardiac operations [12]; penetrating abdominal vascular wounds [13]; emergency abdominal vascular procedures, such as that for ruptured abdominal aortic aneurysms [14]; and penetrating wounds of the deep pelvis [5,15]. Prior to the availability of interventional radiologists and percutaneous detachable balloons [16,17], carotid-cavernous, carotid-jugular, and vertebral-jugular arteriovenous fistulas were also treated with operative placement of intra-arterial balloon catheters
[4,18-22]. In patients with perforations or rupture of the heart, vascular clamps can be used on the atria and finger compression or tamponade on the ventricles, as patients still alive on admission usually have limited injuries. With larger injuries in the ventricles, the previously mentioned inflow occlusion (clamping of the superior and inferior vena cavae) or pharmacologic hypotension may allow for the rapid placement of several deep mattress sutures to close the defect [23]. An alternative approach is to insert a Foley catheter with either a 5-mL or 30-mL balloon into the defect, inflate the balloon, and put traction on the catheter so that the balloon fills the defect [12,23,24]. A cuffed endotracheal tube has been used for the same purpose in one report [25]. A pursestring suture can then be placed around the margins of the defect and pulled up tight as the balloon is deflated and the catheter removed. Besides large cardiac injuries, balloon catheter tamponade is also useful when fragile or inaccessible areas of the heart and central circulation are injured. Examples in the former group would include a thinned-out wall of the
Figure 2. A no. 3 Fogarty balloon catheter containing radiologic dye is visualized in the intracranial carotid artery. The balloon catheter was left in place to prevent recurrence of a high carotidjugular arteriovenous fistula,
right ventricle, the pulmonary artery, and the pulmonary veins. In the latter group, the posterior wall of the right atrium at the caval junctions, posterior wall of the left atrium at the junction with the pulmonary veins, the high posterior wall of the ventricles, and the posterior wall of the ascending aorta and transverse arch are all difficult to occlude with a finger and suture when the heart is still contracting. In each instance, insertion of a balloon catheter may be the quickest and safest technique to control exsanguinating hemorrhage as a decision is reached on the most appropriate technique of cardiorrhaphy or aortorrhaphy. The primary indication for use of balloon catheter tamponade in noncardiac wounds has been the inaccessible location of the injured artery. The carotid artery behind the pharynx or at the base of the skull; the internal maxillary artery and its branches deep in the face; the second, third, and fourth positions of the vertebral artery [26,27]; the internal jugular vein at the base of the skull [11]; and the distal branches of the internal iliac artery in the pelvis are all vessels embedded in or protected by bony structures. Passing a balloon catheter, usually a Fogartytype [28], into the track of a missile or knife wound or into the proximal artery (or distal vein) and then advancing the catheter with periodic inflation will usually allow for control of hemorrhage at the site of injury, despite the lack of classic proximal and distal vascular control. The distal end of the catheter is then attached to a three-way stopcock, the syringe removed, and the catheter left hanging out of the track or one end of the incision. Originally,
THE AMERICAN JOURNAL OF SURGERY VOLUME160 DECEMBER1990 585
FELICIANOET AL
patients in this review were returned to the operating room 3 to 4 days after insertion of the balloon catheter so that deflation could be performed under ideal circumstances. When hemorrhage was noted to be controlled in all instances in the first several patients, subsequent deflations were performed in the surgical intensive care unit and the catheter was removed without a return to the operating room. A n alternative approach is to fold the end of t h e proximal catheter, tie it in this position, and leave it in place, as was described for the patients with cervical arteriovenous fistulas [4]. A difficult problem is the development of ipsilateral cerebral ischemia after balloon occlusion of the high cervical internal carotid artery. If this diagnosis is obvious based on an intraoperative or postoperative electroencephalogram or postoperative computed tomographic scan, rapid performance of an extracranial-intracranial bypass may prevent or reverse neurologic deficits [29]. The one death in this series might have been prevented by such an aggressive approach to cerebral revascularization. The dismal results with operative placement of balloon catheters to control high cervical arteriovenous fistulas in this review suggest that radiologic placement under fluoroscopic control is a better approach. In truth, the patients reported underwent operations in 1980 and 1982, and radiologic techniques have been utilized in our institution since that time, much as in other centers [30]. If operative placement of a balloon catheter to control a high cervical arteriovenous fistula becomes necessary, admittedly an unlikely event in 1990, the technique is as follows: the proximal artery is ligated, the balloon catheter is passed into the distal artery beyond the tie, and the balloon is inflated when the measured level of the fistula is reached or the bruit is noted to disappear after inflation. Inflation of the balloon with radiologic dye will allow for easier visualization on subsequent arteriograms, as previously noted. A n intraoperative cerebral arteriogram is then performed through the contralateral carotid or vertebral artery to verify that retrograde flow through intracerebral collaterals is not keeping the fistula open above the balloon. Intraoperative electroencephalographic monitoring will help determine the need for an ipsilateral extracranial-intracranial bypass if a carotid arteryjugular vein fistula is being occluded. An appropriately placed balloon can be left in place permanently by fixing the proximal end of the catheter in a folded position as described [4]. Balloon catheter tamponade, which has now been performed for 30 years, remains a useful technique in selected patients with hemorrhage from cardiac or vascular injuries that are difficult to control with standard techniques. When exsanguinating hemorrhage from a large defect or a fragile or inaccessible cardiac or vascular structure is encountered, balloon catheter tamponade should be used.
REFERENCES 1. Taylor H, Williams E. Arteriovenous fistula following disk surgery. Br J Surg 1962; 50: 47-50.
586
2. Pearce CW, McCool E, Schmidt FE. Control of bleeding from cardiovascular wounds: balloon catheter tamponade. Ann Surg 1966; 166: 257-9. 3. Foster JH, Morgan CV, Threlkel JB. Proximal control of aorta with a balloon catheter. Surg Gynecol Obstet 1971; 132: 693-4. 4. Binkley FM, Wylie EJ. A new technique for obliteration of cerebrovascular arteriovenous fistulae. Arch Surg 1973; 106: 5247. 5. Sheldon GF, Winestock DP. Hemorrhage from open pelvic fracture controlled intraoperatively with balloon catheter. J Trauma 1978; 18: 68-70. 6. Reilly JJ Jr, Caparosa R J, Latchaw RE, Sheptak PE. Aberrant carotid artery injured at myringotomy. Control of hemorrhage by a balloon catheter. JAMA 1983; 249: 1473-4. 7. Belldn M, Dunton R, Crombie HD Jr, Lowe R. Preoperative percutaneous intraluminal balloon catheter control of major arterial hemorrhage. J Trauma 1988; 28: 548-50. 8. Feliciano DV, Bitondo CG, Mattox KL, et al. Civilian trauma in the 1980s. A 1-year experience with 456 vascular and cardiac injuries. Ann Surg 1984; 199: 717-24. 9. Feliciano DV, Mattox KL. Indications, technique, and pitfalls of emergency center thoracotomy. Surg Rounds 1981; 4: 32-40. IO. Feliciano DV. Managing peripheral vascular trauma. Infect Surg 1986; 5: 659-69, 682. 11. Cohen IK, Chretien PB. Control of hemorrhage from the internal jugular vein by a balloon catheter. Surg Gynecol Obstet 1973; 136: 791-2. 12. Dube P. Control of massive cardiac haemorrhage by balloon catheter. Thorax 1973; 28: 399-400. 13. Smiley K, Perry MO. Balloon catheter tamponade of major vascular wounds. Am J Surg 1971; 121: 326-7. 14. Hyde GL, Sullivan DM. Fogarty catheter tamponade of ruptured abdominal aortic aneurysms. Surg Gynecol Obstet 1982; 154: 197-9. 15. Owen DR, Hodgson PE. Control of hemorrhage following missile wound to the pelvis. J Trauma 1980; 20: 906-8. 16. Barrow DL, Fleischer AS, Hoffman JC. Complications of detachable balloon catheter technique in the treatment of traumatic intracranial arteriovenous fistulas. J Neurosurg 1982; 56: 396-403. 17. Miller RE, Hieshima GB, Giannotta SL, Grinnell VS, Mehringer CM, Kerin DS. Acute traumatic vertebral arteriovenous fistula: balloon occlusion with the use of a contralateral approach. Neurosurgery 1984; 14: 225-9. 18. Proto DJ, Hanbery JW. Intraluminal occlusion of a carotidcavernous sinus fistula with a balloon catheter. Technical note. J Neurosurg 1971; 35: 237-42. 19. Scott TE Jr, Tweed CG, Hollis BF, Eardley K. Management of traumatic carotid cavernous fistula using the Fogarty catheter technique. J Trauma 1980; 20: 610-13. 20. Debrun G, Lacour P, Vinuela F, Fox A, Drake CG, Caron JP. Treatment of 54 traumatic carotid-cavernous fistulas. J Neurosurg 1981; 55: 678-91. 21. Goodman SJ, Hasso A, Kirkpatrick D. Treatment of vertebrojugular fistula by balloon occlusion. Case report. J Neurosurg 1975; 43: 362-7. 22. Fairman RM, Grossman RI, Goldberg HI, Kivuls J, Perloff LJ. A new approach to the treatment of vertebral arteriovenous fistulas. Surgery 1984; 95: 112-5. 23. Trinkle JK, Toon RS, Franz JL, Arom KV, Grover FL. Affairs of the wounded heart: penetrating cardiac wounds. J Trauma 1979; 19: 467-72. 24. Arbuln A, Thorns NW. Control of bleeding from a gunshot wound of the inferior vena cava at its junction with the right atrium by means of a Foley catheter. J Thorac Cardiovasc Surg 1972; 63: 427-9: 25. Oelert H, Dalichau H, Borst HG. The use of a cuffed endotracheal tube for repair of operative injuries to the atria. Ann Th0rac Surg 1974; 18: 102-3. 26. Meier DE, Brink BE, Fry WJ. Vertebral artery trauma. Acute
THE AMERICAN JOURNAL OF SURGERY VOLUME 160 DECEMBER 1990
BALLOONCATHETERTAMPONADEIN CARDIOVASCULARWOUNDS
recognition and treatment. Arch Surg 1981; 116: 236-9. 27. Golueke P, Sclafani S, Phillips T, Goldstein A, Scalea T, Duncan A. Vertebral artery injury--diagnosis and management. J Trauma 1987; 27: 856-65. 28. Fogarty T J, Cranley J J, Krause R J, Strassner ES, Hafner CD. A method for extraction of arterial emboli and thrombi. Surg Gynecol Obstet 1963; 116: 241-4. 29. Gewertz BL, Samson DS, Ditmore QM, Bone GE. Management of penetrating injuries of the internal carotid artery at the base of the skull utilizing extraeranial-intracranial bypass. J Trauma 1980; 20: 365-9. 30. Sclafani SJ, Panetta T, Goldstein AS, et al. The management of arterial injuries caused by penetration of zone III of the neck. J Trauma 1985; 25: 871-81.
DISCUSSION Gilbert S. Campbell (Little Rock, AR): With the tremendous experience and contributions that Dr. Feliciano and his associates have made through the years in trauma, to have utilized this particular procedure only 10 times in 10 years underscores its infrequent need. The balloon-tipped catheter tamponade technique is a lifesaving measure rarely used. Dr. Feliciano, you stated in the manuscript that the balloon-tipped catheter was left permanently in one patient. How long a period of time did you mean by "permanently?"
Richard E. Ward (Sacramento, CA): Were heparin or anticoagulants used in these patients, especially those with carotid balloon tamponade? David V. Felieiano (closing): Dr. Campbell, if the injured artery is in an inaccessible location, these catheters can be folded over and tied in that position. The balloon will stay inflated in the vessel. In most of the patients in this series, the balloons were simply deflated 3 to 4 days later in the operating room or in the surgical intensive care unit because we became convinced that bleeding would not recur with deflation. In two patients who required balloon insertion to control hemorrhage and in one patient with a balloon inserted to close a high carotid-jugular fistula, the balloons have never been removed. Dr. Ward, we did not give heparin to any patients in this series. I am not sure that this would prevent the complication of ipsilateral cerebral infarction. In a small subset of patients, the circle of Willis is not going to protect these patients if one carotid artery is occluded. The lesson in this series is that surgeons should keep this technique in mind. There are few places in the body where one of these small balloon catheters cannot be placed. If the balloon is left inflated for 3 to 4 days, the arterial tree thromboses around it, and there will be no further hemorrhage.
THE AMERICAN JOURNAL OF SURGERY VOLUME160 DECEMBER 1990 587