LAUSR

OBJECTIVE Little is known about the effectiveness of lumbar drainage(LD) in the treatment of delayed or recurrent cerebrospinal fluid(CSF) leaks. We report our institutional experience and the effectiveness of LD in the management of the delayed or recurrent CSF leaks. METHODS Between January 2014 and December 2018, a total 21 patients with delayed or recurrent CSF leaks were enrolled in the research. All patients were managed with conservative treatment for 48 hours, LD was prescribed if CSF leaks still exists more than 48 hours. If LD failed, endoscopic endonasal surgery(EES) was performed as soon as possible. Medical records were collected to analyse the effectiveness of LD. RESULTS Among 21 patients, 4 patients resolved with conservative treatment, 17 patients were treated by CSF diversion via LD. The total cure rate of LD was 9/17(52.9%). The cure rate was no statistically significant differences (relative risk RR=3.33; Fisher's Exact Test P=0.131) between traumatic group (8/12, 66.7%) and transsphenoidal surgery group(1/5, 20.0%). During the follow-up time, no recurrent of CSF leaks was observed. CONCLUSIONS The cure rate of LD in the delayed or recurrent CSF leaks was lower than initial treatment with LD. There is a tendency that the cure rate of traumatic group is higher than transsphenoidal surgery group. EES can be used as a remedial treatment for patients with LD treatment failed.