CASE REPORT Tracheo-Innominate Artery Fistula After Tracheostomy Leonardo Kapural, MD*, Juraj Sprung, MD, PhD†, Ivo Gluncic, MD‡, Miranda Kapural, MD*, Simon Andelinovic, MD§, Dragan Primorac, MD§, and Peter K. Schoenwald, MD† *Division of Anesthesiology and Critical Care Medicine and †Department of General Anesthesiology, The Cleveland Clinic Foundation, Cleveland, Ohio; and Departments of ‡Otorhinolaryngology and §Pathology, Split School of Medicine, Split, Croatia F istula formation between the trachea and the in- nominate artery is a rare complication of trache- ostomy (1). The survival rate in patients who develop bleeding from a tracheo-innominate artery fistula (TIF) has been reported as 14.3%, and only patients who received immediate surgical treatment survived (2). Of those patients who develop a TIF, 78% do so within the first 3 wk after tracheostomy (3). One of the proposed mechanisms of fistula formation is mucosal necrosis due to pressure caused by the elbow, tip, or cuff of the tracheostomy tube (2). Clinical presentations and treatment of TIF have been described mainly in the surgical literature (1–7). However, because anesthesiologists may be involved in treating this emergency, they must be familiar with the therapeutic steps. We present a patient who de- veloped a TIF and died as a consequence of massive hemorrhage into the tracheobronchial tree with as- phyxia. We discuss potential preventative measures that should be followed to decrease the probability of formation of a tracheo-arterial fistula, as well as im- portant diagnostic and therapeutic steps the anesthe- siologist must take in managing this severe tracheos- tomy complication. Case Report A 22-yr-old, 80-kg man (height 175 cm) suffered a severe head injury in a motor vehicle accident. On arrival at the hospital, his trachea was intubated. A computed tomogra- phy scan of the head and cervical spine showed an epidural hematoma, fracture of the second cervical vertebrae, multi- ple fractures of the occipital bone, and dislocation of the left atlantoaxial joint without spinal canal compromise. Suboc- cipital osteoclastic craniotomy and evacuation of the epi- dural hematoma were performed immediately, and the neck was conservatively stabilized. Because there was no dislo- cation of the cervical vertebrae or risk to the cord, further surgical intervention was not deemed necessary. Postoper- atively, the patient was transferred to the intensive care unit, where his breathing was maintained with mechanical ven- tilation. He was comatose for the first 5 postoperative days (PODs). On POD 6, he regained some spontaneous move- ment of his left upper and lower extremities. On POD 9, the patient began to communicate and was aware of his surroundings. Because his breathing remained ventilator- dependent, a tracheostomy was performed on POD 11 at the level of the second, third, and fourth tracheal rings with a transverse incision. A low-pressure size 8 cuffed tracheos- tomy tube was placed. On POD 16, the patient became febrile, and the urine culture was positive for pseudomonas. Antibiotic therapy was begun, after which the patient be- came afebrile on POD 17; on POD 26, antibiotics were dis- continued. On POD 30, minimal bleeding from the trache- ostomy was noticed. The tracheal cuff was deliberately overinflated and the bleeding temporarily ceased. However, 1 h later, while being transported to the operating room (OR) for surgical exploration, massive tracheal bleeding occurred. Digital compression of the bleeding site through the trache- ostomy opening against the sternum was attempted without success, and cardiac arrest ensued. Immediately after the bleeding started, the tracheostomy tube was replaced with an oral endotracheal tube. Despite all resuscitation attempts, including replenishment of intravascular volume with col- loid solution (hetastarch), lactated Ringer’s solution, and whole blood, as well as cardiopulmonary resuscitation ef- forts (chest compression and epinephrine), the patient died. An autopsy revealed massive blood aspiration into the lungs from a large TIF. Histopathologic examination of the fistula revealed necrosis at the tracheal opening, which suggests that a necrotic ischemic process started at the trachea and progressed toward the innominate artery. Discussion TIF is a life-threatening complication of tracheostomy that usually presents with acute and massive tracheal bleeding. Without prompt surgical intervention, the outcome of this complication is grave. Therefore, a high index of suspicion should be maintained in any patient with tracheostomy and subsequent hemopty- sis (1). Premonitory minimal tracheal bleeding and pulsation of the tracheostomy tube synchronous with the heart beats have been reported as warning signs of Accepted for publication December 23, 1998. Address correspondence and reprint requests to Juraj Sprung, MD, PhD, Division of Anesthesiology and Critical Care Medicine, The Cleveland Clinic Foundation, 9500 Euclid Ave., E31, Cleveland, OH 44195. ©1999 by the International Anesthesia Research Society 0003-2999/99/$5.00 Anesth Analg 1999;88:777–80 777