Discussion

The first mention of balloon catheter tamponade used clinically in the American surgical literature was by William E Schroeder from Cook County Hospital in 1906.1 He described a patient with a gunshot wound of the “thickest portion of the liver” (right lobe) from a 44-caliber revolver in 1900. To control “profuse hemorrhage” from the liver at operation, he inserted a rubber bag into the track of the missile with a firm rubber rod next to it. After inflating the bag the hemorrhage was controlled and gauze packing was then placed at the inferior border of the liver, as well. The “ends of the gauze packing, tube, and rubber bag were brought out at the upper angle” of the right pararectus incision. The gauze packing, tube, and rubber bag were removed at 7 days, and the “patient made a complete recovery at the end of four weeks.”

Experimental balloon catheter occlusion of the abdominal aorta to control intra-abdominal hemorrhage was first suggested in a research article published from the American College of Surgeons’ Surgical Forum in 1953.2 3 Less than 1 year later, the late Major General Carl W Hughes (a lieutenant colonel at the time of the Korean War) described the insertion of a no. 10 French balloon catheter through the femoral artery up into the abdominal aorta at the level of the diaphragm in two casualties who were moribund from shock prior to laparotomy. In the first patient the blood pressure rose from “none” to 110/70 after inflation of the intra-aortic balloon.4

One of the earliest subsequent uses of clinical balloon catheter tamponade was not described until 62 years after Schroeder’s patient from 1900. H Taylor and B Williams5 from The London Hospital used balloon catheter tamponade as an intraoperative maneuver to control venous hemorrhage during closure of a common iliac arteriovenous fistula 2 years after a “disk operation.”

In 1966, Pearce et al6 demonstrated the value of balloon catheter tamponade in patients with penetrating wounds to the chest. He described the use of temporary balloon tamponade in two patients with penetrating wounds to the right pulmonary artery and inferior vena cava/right atrium, respectively.

During the past six decades, there has been an extraordinary increase in the indications for balloon catheter tamponade, particularly in injured patients (box 1).4–20 These multiple indications can be summarized as follows:

• Temporary intraoperative insertion of a balloon catheter which is then inflated into a visible vascular or cardiac wound to allow for placement of sutures in a “dry field.”

• Balloon catheter tamponade inserted intraoperatively and then maintained for 48 to 72 hours to control hemorrhage from an “inaccessible” location. The corollary to this statement is that an intravascular balloon occluding the arterial or venous flow always causes thrombosis of the vessel.

• Permanent placement of a detachable balloon to occlude a cerebral arteriovenous fistula or carotid-cavernous sinus fistula.

In terms of equipment, Fogarty balloon catheters are available in 2F to 7F sizes for arteries and in 6F to 8F for veins. These catheter balloon sizes are ideal for small missile or stab tracks leading to the internal carotid artery at the base of the skull, or for direct insertion into the cervical, thoracic, abdominal (not the aorta or inferior vena cava), pelvic, or peripheral vessels. Foley catheters are available with 5 mL or 30 mL balloons. These larger-sized balloons are ideal for insertion into gaping facial, cervical, subclavian, axillary, groin, and popliteal wounds where exsanguinating hemorrhage is occurring from an unknown vessel. A third option used for intrahepatic tamponade was first described by Morimoto et al15 from the University of Sao Paulo, Brazil, in 1987. A Robinson urethral catheter is inserted into a Penrose drain; the Penrose drain is tied tightly down at its ends onto the Robinson catheter, and the distal end of the Robinson catheter is occluded with ties as well. Once this device is in place in the parenchyma of the bleeding liver, an injection of fluid into the Robinson catheter will inflate the Penrose drain into a long balloon for tamponade.

In the modern era, the availability of endovascular stents and stent grafts has decreased the need for postoperative or permanent balloon occlusion. After balloon tamponade has controlled the hemorrhage, an early CT arteriogram will define the location, character, and magnitude of the vascular lesion. This may require temporary deflation of the balloon in a controlled location such as a hybrid operating room. Also, when wounds at the base of the skull are managed with balloon catheter tamponade, the internal carotid artery (zone III) is thought to be the vessel being occluded. If a CT arteriogram demonstrates that crossover cerebral flow is inadequate to perfuse the ipsilateral brain, conversion of the occluding balloon to an endovascular stent or stent graft is necessary.

In the largest single center civilian review to date published in 2011, 21 injured patients had “brief” (<6 hours; essentially intraoperative) and 23 patients had “prolonged” (>6 hours) insertion of a balloon catheter.21 Of interest, the mean indwelling times for balloon catheters in the “prolonged” group were as follows: 78 hours for carotid tamponade at the base of the skull; 53 hours for intrahepatic tamponade; and 31 hours for iliac tamponade. The most common locations for balloon catheter tamponade from 1998 to 2009 in this review were the “face/pharynx/base of skull” in 12 patients (survival 8%) and “liver” in 12 patients (survival 67%).