Management of cerebrospinal fluid leak associated with craniomaxillofacial trauma

Purpose: The management of persistent, post-traumatic cerebrospinal fluid (CSF) rhinorrhea and otorrhea remains a surgical challenge. Repair of CSF leaks has evolved from that of an intracranial approach to one that is primarily extracranial and endoscopic. The purpose of this retrospective analysis is to determine the incidence of persistent CSF rhinorrhea and otorrhea and assess the clinical outcomes of patients presenting to a level I trauma center with posttraumatic CSF leaks who were managed by both surgical and nonsurgical means. Patients and Methods: The records of all patients with basilar skull fractures and/or severe facial trauma presenting to a major level I trauma center from 1991 to 2001 were reviewed. Patients diagnosed with CSF otorrhea or rhinorrhea, who had not undergone an intracranial procedure, elevation of depressed skull fractures, or received a ventriculostomy, were identified and their demographics recorded. For purposes of statistical comparison, patients were divided into 2 groups: leak and no leak. All patients in the leak group were initially observed for a period of 7 to 10 days. Persistent CSF leaks were managed by CSF diversion via lumber drainage for 5 to 7 days. Extracranial repair was performed only if lumbar drainage failed to resolve the leak. Results: Seven hundred thirty-five patients were identified who met the criteria for inclusion in the study. Thirty-four patients (incidence, 4.6%) were identified with CSF leak presenting as otorrhea (n = 25 [75.8%]) or rhinorrhea (n = 9 [26.5%]), which was diagnosed by clinical, laboratory, or radiographic examination (average age, 28.2 years; age range, 2 to 80 years; 23 males and 11 females). All patients in this study experienced successful resolution of CSF otorrhea or rhinorrhea by using a variable combination of observation, CSF diversion, and extracranial repair. There were no complications or cases of meningitis. Twenty-eight patients (84.6%) experienced uncomplicated resolution of the leak without treatment in 2 to 10 days. Persistent CSF leak, defined by drainage greater than 7 days after injury, was identified in 6 patients (incidence, 0.8%), all except 1 who underwent CSF diversion via a lumbar drain for a period of 5 to 10 days. Two of these patients were treated successfully; the remaining 4 patients required surgical procedures. Conclusions: Post-traumatic CSF leaks are uncommon andwill usually resolve without surgical intervention. Successful management in refractory cases often involves a combination of observation, CSF diversion, and/or extracranial and intracranial procedures. (C) 2004 American Association of Oral and Maxillofacial Surgeons.