Minimally Invasive Endoscopic Surgery for Cerebrospinal Fluid Rhinorrhea

The objective of this study is to evaluate the clinical effect of endoscopic surgery for cerebrospinal fluid (CSF) rhinorrhea and discuss the best treatment. Records of 52 patients, ages 19 to 60 years with a mean age of 41.9 years, who were treated by endoscopic surgery in our department between June 2002 and May 2010 were analyzed retrospectively. Eighteen were spontaneous and 34 were traumatic or iatrogenic. Middle turbinate mucosa, fascia lata, and abdominal fat were used in the leak repair. The CSF fistulas were closed by the combined multilayer “underlay,” “overlay,” and “bath-plug” techniques. The cribriform plate was the most common site of the leak (21/52), followed by the sphenoid sinus (17/52), anterior ethmoid sinus (7/52), posterior ethmoid sinus (4/52), and frontal recess (3/52). All cases had only one defect. Defects ranged in size from 5 to 35 mm (mean, 9.7 ± 3.5 mm); 13 cases were >10 mm. Forty-two cases (80.8%) were repaired successfully at the first transnasal endoscopic surgery, six cases succeeded at the second attempt, and four cases succeeded at the third attempt. Ultimately, 100% (52/52) of the leaks were closed. The first-attempt success rate of patients with large size (>10 mm) leaks (61.5%) was significantly lower than that of patients with small size (<10 mm) leaks (87.2%). One patient had postoperative intracranial infection and one patient had postoperative hydrocephalus. No other postoperative complications were observed. All patients were followed-up for 6 months to 3 years (mean, 1.2 years), and no recurrence was founded. Endoscopic closure of CSF rhinorrhea represents a minimally invasive and highly successful procedure. Location and size of the leak and material used to repair the leak affect surgical outcome.