OBJECTIVES/HYPOTHESIS: Trauma and surgery are the most common causes of cerebrospinal fluid (CSF) rhinorrhea. Surgical repair is recommended for patients with CSF leaks that do not respond to conservative measures, traumatic CSF leaks that require transcranial surgery for associated brain injuries, and iatrogenic defects that are discovered intraoperatively. The purpose of our study was to ascertain the outcome after transnasal endoscopic repair of CSF leaks and to identify factors regarding the patient, CSF fistula, and treatment that may influence the results of the repair. METHODS: We performed a meta-analysis of all studies published in English between 1990 and 1999 that reported a minimum of five patients with CSF fistulae that were repaired using an endoscopic approach. We analyzed data that included type of graft and technique used during the repair, surgical complications, the use of packing, and the use of lumbar drains and antibiotics. The success rate was monitored and correlated with the other variables. The meta-analysis database was compared with and added to a database comprising our own patients. RESULTS: Fourteen studies comprising 289 CSF fistulae met the inclusion criteria. Endoscopic repair of CSF leaks was successful in 90% (259/289) of the cases after a first attempt. Seventeen of 30 persistent leaks (52%) were closed after a second attempt. Thus ultimately 97% (276/289) of the leaks were repaired using an endoscopic approach. The success rate of repairs using any of the reported techniques and materials was high and not statistically different. The incidence of major complications such a meningitis, subdural hematoma, and intracranial abscess was less than 1% for each complication. CONCLUSION: The endoscopic approach is highly effective and is associated with low morbidity. The literature supports the endoscopic approach using a variety of techniques and materials for the repair of CSF leaks.
OBJECTIVES/HYPOTHESIS: Trauma and surgery are the most common causes of cerebrospinal fluid (CSF) rhinorrhea. Surgical repair is recommended for patients with CSF leaks that do not respond to conservative measures, traumatic CSF leaks that require transcranial surgery for associated brain injuries, and iatrogenic defects that are discovered intraoperatively. The purpose of our study was to ascertain the outcome after transnasal endoscopic repair of CSF leaks and to identify factors regarding the patient, CSF fistula, and treatment that may influence the results of the repair. METHODS: We performed a meta-analysis of all studies published in English between 1990 and 1999 that reported a minimum of five patients with CSF fistulae that were repaired using an endoscopic approach. We analyzed data that included type of graft and technique used during the repair, surgical complications, the use of packing, and the use of lumbar drains and antibiotics. The success rate was monitored and correlated with the other variables. The meta-analysis database was compared with and added to a database comprising our own patients. RESULTS: Fourteen studies comprising 289 CSF fistulae met the inclusion criteria. Endoscopic repair of CSF leaks was successful in 90% (259/289) of the cases after a first attempt. Seventeen of 30 persistent leaks (52%) were closed after a second attempt. Thus ultimately 97% (276/289) of the leaks were repaired using an endoscopic approach. The success rate of repairs using any of the reported techniques and materials was high and not statistically different. The incidence of major complications such a meningitis, subdural hematoma, and intracranial abscess was less than 1% for each complication. CONCLUSION: The endoscopic approach is highly effective and is associated with low morbidity. The literature supports the endoscopic approach using a variety of techniques and materials for the repair of CSF leaks.