Abstract Date: 5/1/2016
Nathan T. Zwagerman, MD
Paul Gardner (Pittsburgh, PA)
Introduction CSF leaks remain a complication after endoscopic endonasal surgery. In this study, subjects were randomized to receive an immediate postoperative lumbar drain or not with comparison of postoperative leak rates.
Methods The study randomized patients to lumbar drainage or no drain following reconstruction. The inclusion criteria included: 1) extensive arachnoid dissection, 2) dissection into a ventricle or cistern, or 3) dural defect greater than 1cm2. Demographic data, tumor location, defect size, and leak rates were collected. The study was approved by the Institutional Review Board. (PRO010030258)
Results The trial was stopped early (170 patients) due to a significant difference in CSF leak rate between the experimental (drain) and control (no drain) groups. No significant differences were found in the demographic measures between the two groups with an average age of 51.2 years (19-86)and BMI of 31.1. The most significant variable for postoperative leak was not having a drain (p<0.011). A difference in leak rate was found based upon the tumor location (anterior, posterior, suprasellar) (p<0.021). However, when a drain was employed, tumor location was not a significant factor in post-op leak (p<0.507). Defect size was noted to be larger in the group with leaks compared to the control group (6.86 cm2 v. 2.78 cm2, p < 0.076). Thirty-six patients had anterior pathology (olfactory groove or planum). The leak rate was 10% with and 35% without a drain (p<0.11). Fifty patients had posterior pathology (clival). The leak rate was 13% with and 30% without a drain. (p<0.12). Finally, 85 patients had suprasellar lesions. The leak rate was 4.7% with and 9.5% without a drain (p<0.43).
For patients undergoing endoscopic endonasal skull base surgery, lumbar drain placement lowers the rate of postoperative CSF leak. The impact seems to be greatest in patients with large anterior or posterior cranial base defects.
Article ID: AA-34944