Once considered relatively inaccessible, nestling amid cranial nerves, the clivus has become a reasonable target for the surgeon with the advent of new midline ventral approaches. Since Kanaval' described the transoral removal of a bullet from the clivus in 1909, technique and instrumentation have advanced such that transoral surgery2 and maxillotomy3 provide practical access to the extradural clivus with excellent cosmetic results and few complications.
CASE REPORTA previously fit 54-year-old man was admitted to the National Hospital for Neurology and Neurosurgery for removal of a craniocervical chordoma. This lesion occupied the lower 1.5 cm of the clivus, extending around the occipital condyle on the right, and to a much lesser extent on the right. Some tumor extended behind the odontoid peg; the peg itself and the arch of the first cervical vertebra, however, were intact.An "open-door" maxillotomy using a transverse osteotomy at the Le Fort 1 level combined with a midline spit of hard and soft palates was the surgical strategy chosen. Preoperatively a percutaneous gastrostomy was fashioned to permit immediate postoperative feeding. A tracheostomy was inserted at operation. Good exposure of this clivus was obtained by this approach, and radical tumor clearance was effected. The vertebral arteries and their muscular branches were identified and preserved. Closure was meticulous, with fibrin glue and surgical cellulose applied to the tumor defect. The posterior palate was compression wired and the maxilla fixed with mini-plates. The mucoperiostium was closed and hydrocortisone ointment applied to the wound.The patient made an excellent postoperative recovery. Within 5 days, he was taking a light diet orally and was completely self-caring. On the fifth postoperative day, he suffered a massive epistaxis that resulted in this becoming shocked. He initially responded to aggressive fluid resuscitation, and his nasopharynx was packed. Later he described the sensation of a "jet of blood, high up . . . in my nose." The epistaxis recurred and over the next 36 hours this man would require 6 units of blood transfusion. His clotting profiles and liver function were within normal ranges throughout. Pernasal exploration under anesthesia failed to identify the site of his bleeding.153