Spontaneous cerebrospinal fluid leak repair: a five-year prospective evaluation.

OBJECTIVES/HYPOTHESIS Mounting evidence indicates the majority of spontaneous cerebrospinal fluid (CSF) leaks are associated with intracranial hypertension. The objectives of the current study were to assess outcomes regarding spontaneous CSF leaks focusing on premorbid factors, surgical technique, and management of intracranial pressure. STUDY DESIGN Prospective cohort. METHODS Prospective evaluation of patients with spontaneous CSF leaks was performed. Data regarding demographics, nature of presentation, body mass index (BMI), location and size of defect, intracranial pressure, clinical follow-up, and complications were collected. RESULTS Over 5 years, 46 patients (average age, 51 years) with 56 spontaneous CSF leaks were treated by a single otolaryngologist. Twenty-one subjects presented with recurrence of their CSF leak following previous endoscopic and/or open approaches by other physicians. Obesity was present in 78% of individuals (average BMI, 35.6). Fifty-two CSF leaks (93%) were successfully repaired at first attempt. With secondary repair, all CSF leaks were closed at last clinical follow-up (average, 93 weeks). Three patients developed late failures (>2 months), with one recurrence at a distinct location from the primary site at 8 months postprocedure (associated with ventriculoperitoneal shunt failure). Opening pressures via lumbar puncture averaged 24.3 ± 8.3 cm H2 0, which increased significantly to 32.3 ± 9.0 cm H2 0 (P < .0001) following closure of the skull base defect(s). Management of intracranial hypertension included acetazolamide (n = 23) or permanent CSF diversion (n = 19, including five revisions of failed preexisting shunts). CONCLUSIONS Although spontaneous CSF leaks have the highest recurrence rate of any etiology, prospective evaluation demonstrates high success rates with control of intracranial hypertension.