Background: In cerebrospinal fluid (CSF) leakage endoscopic repair, lumbar drains (LDs) were used in an attempt to increase success rates. To critically assess the relationship between use of LDs and recurrent leaks, we embarked on this randomized prospective study.
Methods: Patients undergoing CSF leak repair between 2000 and 2012 were randomized into two groups: 75 patients were managed without LDs and in the other group of 75 patients an LD was always placed. Different parameters were analyzed to identify their relationship with failures: occurrence of increased intracranial pressure (ICP), body mass index, smoking habits, existence of diabetes, chronic corticosteroid use, previous sinus surgery, etiology (traumatic, iatrogenic, or spontaneous leak), site, and size of the CSF leak.
Results: Success rate was 93% for the whole group. Patients managed with LDs attained 95% success rate and those without LD attained 92%; the difference is not significant (p = 0.2). The only factor predictive of recurrence is increased ICP: 77% success rate versus 97% for traumatic leaks and 96% for iatrogenic leaks. Recurrence rates were identical in the two groups with increased ICP, regardless of the use of a LD (23%).
Conclusion: In this study, success rates of CSF repair were not associated with the use of LDs. However, the small number of cases with high success rates precludes appropriate statistical analysis.